Paper no. 2008/13

RIS and Developing Countries: Linking firm technological capabilities to

regional systems of innovation

Padilla, Ramon (Ramon.PADILLA@cepal.org)

United Nations Economic Commission for Latin America and the Caribbean (ECLAC), Mexico

Vang, Jan (jv@engsci.aau.dk)

Copenhagen Institute of Technology, Aalborg University

Chaminade, Cristina (cristina.chaminade@circle.lu.se) CIRCLE, Lund University

This is a pre-print version of a chapter that will be published in [Lundvall, B-A, Joseph, KJ., Chaminade, C. and Vang, J. (Eds), Handbook of Innovation systems and developing countries. Building domestic capabilities in a global setting] http://www.e-elgar.co.uk ©: [2009] Published by Edward Elgar. All rights reserved. Citations to and quotations from this work should reference that publication. If you use this work, please check that the published form contains precisely the material to which you intend to refer.

Centre for Innovation, Research and Competence in the Learning Economy (CIRCLE) Lund University

P.O. Box 117, Sölvegatan 16, S-221 00 Lund, SWEDEN http://www.circle.lu.se/publications

ISSN 1654-3149

WP 2008/13 RIS and Developing Countries: Linking firm technological capabilities to regional systems of innovation Padilla, Ramon; Vang, Jan and Chaminade, Cristina Abstract This chapter explores the role of regional innovation systems supporting capability building

among indigenous SMEs in two different RIS in Mexico. It explicitly attempts at testing the

validity of the underlying assumptions in RIS literature in the context of developing countries,

such as the importance of interactive learning with foreign subsidiaries, universities and

research centers or among firms in the system of innovation. Empirically, it proposes a new

method for comparing capabilities at organizational and system level. The paper shows that

regional innovation systems in developing countries share central characteristics with RIS in

developed countries and, in that respect, the approach is valid for the analysis of RIS in

developing countries. The analysis also highlights the importance of local conditions for

catching-up and development. The same industry in the same country might perform very

differently depending on the characteristics of local systems.

Keywords: Regional Innovation System, Mexico, developing countries

Disclaimer: All the opinions expressed in this paper are the responsibility of the individual

author or authors and do not necessarily represent the views of other CIRCLE researchers.

RIS and Developing Countries: Linking firm technological capabilities to regional systems of innovation Padilla, Ramon, Vang, Jan and Chaminade, C. 1. Introduction In recent years, there has been an increasing interest in the rapid growth of certain regions and industries in developing countries. The new global landscape characterised by rapid technological development and change, economic globalization, new business strategies and deregulation has opened new windows of opportunities for upgrading and growth in developing countries (Archibugi and Pietrobelli, 2003). A ‘handful’ of regions in the developing world have already managed to utilize the opportunities that the new global landscape provides to accumulate technological capabilities and occasionally even become specialised hubs in global knowledge networks (Chaminade and Vang, 2006a; Asheim et al, 2006). While some countries and regionsi show clear signs of being on the right track, others – especially in Africa and parts of Latin America - are falling behind in terms of upgrading, growth, unemployment and poverty (Kaplinskly, 2005). There appears to be no ‘best practice’-lessons that can be learnt from the successful regions as they have followed highly diverse industrialization, development and upgrading paths. The countries and regions have also different sizes (i.e. home markets), human, social, financial and physical endowments and follow different, partly path dependent, policy intervention strategiesii. The analysis of the existing experiences is also limited by the absence of systematic comparative analysis of different regions and industries across the globe. Hitherto, the existing literature has tried to explain differences in the performance of the various regions focusing on the the strategy of particular firms, the vertical and horizontal links in the clusters, the human capital endowment or the orientation to export markets and the role of the state. Several studies in both developed and developing countries link successful upgrading to the exploitation of agglomeration economies (Scott and Garofoli, 2007). In this context, scholars, consultants and policy-makers have increasingly acknowledged the importance of analyzing and constructing regional innovation systems (RIS) as a means for facilitating catching up processes in firms in developing countries (Asheim et al, 2006). This has spurred an invaluable stream of literature re-theorizing, re-conceptualizing and adapting the ideas behind RIS and related concepts (i.e. clusters) to the specificities and contingencies of developing countries (Lundvall et al 2006, Chaminade and Vang 2006, Pietrobelli and Rabelotti 2005 and 2006, Yeung 2006, Vang 2006 and Asheim, Guiliani et al. 2005, Schmitz, 2006). Yet, while this stream of research has provided valuable insights into the role of RIS in supporting upgrading in firms in developing countries, there are still significant theoretical and methodological gaps. Theoretically, the exiting literature continues to be rather generic, ignoring the specificities of the firms located in the RIS in developing countries (their strategy and role in the value chain) or the specific stage in the evolution of the RIS. Methodologically, it is also suggested that there is a need to move from individual cases to the systematic comparison of regions and develop a systematic and

2

rigorous method to study the dynamics of regional innovation systems in developing countries in a comparative perspective (Chaminade and Vang, 2006b). This chapter aims at contributing to this stream of research by investigating the need to adapt RIS to the specificities of developing countries and proposes a method to systematically analyse and compare the performance of RIS in supporting upgrading of firms in developing countries. In this respect, the chapter contributes to the existing literature by contextualizing the discussion of the importance of different interactions within the RIS to the type of firm (i.e. its technological capabilities and its position in the global value chain). Additionally, the chapter proposes a new methodology to conduct comparative analysis on the role of RIS supporting capability building among (indigenous) firms. For doing so, we will focus on the analysis and comparison of two regions in Mexico with a strong presence of firms in the electronics industry (Jalisco and Baja California). By applying the framework developed in the first part of the chapter, we will analyse the differences in the role of two RIS supporting the acquisition of technological capabilities by the firms located in the region. The chapter draws on original data collected on-site in two Mexican regions in 2004. The reminder of the chapter is structured as follows. First, the theoretical section is presented. This section synthesizes and critically revises the fragmented bits of the literature on RIS and upgrading of firms in developing countries with the aim of deriving specific testable hypothesises. The section opens with a short general introduction. This is followed by a methodological section that introduces how qualitative and quantitative data sources are integrated and presents the specific measures used for testing the hypotheses. Following we apply the proposed method to the comparison of two RIS in Mexico – Jalisco and Baja California - where we test the hypotheses. The chapter is concluded by discussing its contribution as well as further (methodological) challenges and implications for policy makers. 2. Regional Innovation Systems in the literature 2.1. Concept and components of a Regional Innovation System This chapter departs from the Regional Innovation Systems (RIS) approach. RIS’s are defined as a “constellation of industrial clusters surrounded by innovation supporting organizations” (Asheim and Coenen, 2005). Industrial clusters refer to the geographic concentration of firms in the same or related industries (Porter, 1998; Pietrobelli and Rabellotti, 2004; for a critique, see Martin and Sunley, 2003). The concept of RIS was developed on the basis of and inspired by successful regions and clusters as Silicon Valley (Cohen and Fields 1998; Saxenian, 1994), Baden Württemberg (Staber, 1996) and the Third Italy (Beccatini, 1990; Piore and Sabel, 1984). As such, most of the literature on regional innovation systems reflects the traits and characteristics of the developed world. It has even been suggested that the so-called Holy Trinity does not reflect the

3

developed world as such but ‘outlayer’ regions (Intarakumnerd and Vang 2006). Across the different interpretations RIS approaches stress the systemic dimensions or propensities of the innovation process; being the dynamic interaction between the different components of the system, that is individuals, organizations and institutions and their interactions (i.e. viewing innovation as an interactive process; not a linear one). Conceptually RIS are conceived as ex post rationalizations of the aforementioned success cases, that is, what the literature considers to be a well-functioning system is mainly a generalization of the successful cases of Silicon Valley, Baden Württemberg or the Third Italy. RIS in developing countries can be understood as ex ante constructions of RIS (Intarakumnerd and Vang 2006, Lundvall et al, 2006), where in most cases we can only find some of the elements of an emergent RIS. RIS in developed and developing countries face fundamentally different theoretical challenges as they are embedded in different institutional frameworks (typically weak indigenous formal institutions and strong international bodies) and temporal specificities (catching up as opposed to first movers) and – often – rely on capital and knowledge originating not just outside the sub-national regions borders but outside the country (Amin, 2004, Loebis and Schmitz, 2005; Pietrobelli and Rabellotti, 2006; Schmitz 2006). The lack of local knowledge resources in RIS in developing countries forces the indigenous firms to rely much more on TNCs as providers of knowledge and capital (Pietrobelli and Rabellotti, 2006; Schmitz 2006; Vang and Asheim, 2006). In this context, a critical question is under which conditions RIS in developing countries can support upgrading and the acquisition of technological capabilities by indigenous firms. In this sense, it is useful to distinguish between upgrading in firms and upgrading of the whole system. Firm upgrading is defined as the capacity of a firm to innovate and/or increase the value added of its products and processes (Humphrey and Schmitz, 2002, Chaminade and Vang, forthcoming).Similarly to firms, a regional system possesses technological capabilities, understood as “knowledge and skills embedded in individuals, organizations and institutions located in a geographically-bounded area and conducive to innovative activity” (Padilla-Perez, 2006, p. 69). Regional “systemic” innovation capabilities are not simply the sum of individual firm-level technological capabilities developed in isolation (Lall, 1992). A region embeds many systemic elements external to the firm, which influence its technological competence and growth (Cooke et al., 1997; Howells, 1999; Evangelista et al., 2002; Iammarino, 2005). Meso-level capabilities cannot thus only be conceptualized as the sum of the technological capabilities of the innovation-oriented organisations in the region; their interactions are considered crucial (von Tunzelmann, 2006). Nevertheless, the development of regional capabilities shares most of the features of firm-level capabilities in that regional learning is a long, uncertain and costly process, displaying high path-dependence and cumulativeness. To study regional systemic innovation capabilities, the basic elements need to be identified: the components, their attributes or functions, and their relationships.iii (Regional) innovation systems can be generally decomposed into two sub-systems (Padilla-Perez, 2006, Carlsson, 2006): the firms within a cluster. Asheim and Coenen, (2005) refer to these as constituting the knowledge exploitation subsystemiv) and the

4

knowledge support organizations (often referred to as the knowledge exploration subsystem) in which universities (universities, technical colleges, and R&D institutes), technology transfer agencies, vocational training organizations, business associations, finance institutions, etc are included (see Figure 1) (Asheim and Coenen, 2005). The latter are considered central in creating (upgrading, innovation, unlearning and processes of creative destruction), importing, adapting and retaining knowledge in the RIS. The interaction with the knowledge exploration subsystem is considered critical in the realm of radical innovations while incremental innovations are often attributed to in-house activities or seen as emerging from the daily interactions between firms, users and subcontractors (Lundvall, 1988, Piore and Sabel, 1984, Von Hippel 1988). Figure 1

Knowledge Exploration and Support Subsystem (universities, tech transfer offices, etc)

Knowledge Exploitation Subsystem (business, cluster)

Intra-RIS Knowledge exchange / Interactive Learning

Extra-RIS Knowledge exchange / Interactive Learning(e.g. scientific communities, (inter)nationalresearch collaboration

Extra-RIS Knowledge exchange / Interactive Learning(e.g. global value chains, transnational corporation)

Knowledge Exploration and Support Subsystem (universities, tech transfer offices, etc)

Knowledge Exploitation Subsystem (business, cluster)

Intra-RIS Knowledge exchange / Interactive Learning

Extra-RIS Knowledge exchange / Interactive Learning(e.g. scientific communities, (inter)nationalresearch collaboration

Extra-RIS Knowledge exchange / Interactive Learning(e.g. global value chains, transnational corporation)

GovernancePolicy Institutions

RIS

Source: Coenen (2006) and Vang et al (forthcoming) Well-functioning RIS are characterised by the high level of technological capabilities of the organizations in the system, the large scale and scope of interactions between these two subsystems, as well as the intensity, density and breadth of the outward flows with the rest of the world. That is, RIS should not be reduced to interactions within the local actors, but also embracing knowledge flows with other organizations located outside the region (Giuliani et al, 2005; Vang and Chaminade, 2006, Chaminade and Vang, 2006). The scope of the interactions strongly influences by the institutional framework. The institutions (the rules, norms and values) are seen as the regulating devices ordering, in a non-deterministic way, the behaviour of the actors and their interaction in the RIS. Especially central is that institutions reduce uncertainty as they induce behavioural regularity and thus shape or influence the morphology of the relationships (r) (i.e. the balance between traded and un-traded interactions - across the agents/components (c) in the system. The institutions can be formal (for example, intellectual property rights) or informal (norms and values).v

Finally, the system of innovation can be shaped by regional innovation policy – not to mention other policy topics as industrial policies and sound macro economic polices. Innovation policy should not be conflated with science and technology policy as it is

5

broader and encompass competency and capacity building as well as policies targeting knowledge dissemination, attraction, adoption and retention. Yet, as emphasized by much of the development-literature (i.e. focus on the (post)Washington consensus) and underscored by Isaksen (2003) the functioning of the RIS is also influenced by policies designed and implemented outside the boundaries of the region, for example through national science and technology policy and central decisions about the extent and level of regional administrative devolution. Generally speaking, RIS policy is argued to improve the performance of the regional innovation system by supporting the creation, acquisition and retention of technological capabilities and the diffusion of relevant knowledge among the actors embedded in the system. But the objectives and the instruments that might be used for each RIS as well as the degree of intervention of the government in the regional system of innovation varies significantly across regions (Asheim and Isaksen, 2002, Vang and Chaminade, 2006). The ability of a firm or other organization to tap into, maintain, acquire, use and retain the knowledge available in the system is a function of its absorptive capacity (Cohen and Levinthal, 1990, Giuliani and Bell, 2005). A firm’s absorptive capacity is a function of its prior internal knowledge– being tacit or codified and its learning ability (investments in developing new knowledge) as well as the institutional setting (Vang and Chaminade, 2006). In other words, it is a function of its (investment in) technological capabilities, being defined as the knowledge and skills needed to acquire, use, adapt, improve and create technology (Padilla-Perez, 2006, p. 13). The existing literature widely acknowledges the importance of technological capabilities for long-term, upgrading- and innovation-based growth. Such capabilities are essential both to create new technologies and to effectively use technologies developed in other countries, as well as to adapt and improve existing technologies. The development of capabilities has the following characteristics. First, technology within firms is accumulated in a gradual and costly manner with firms engaging in the long and continuous efforts of incremental learning (Hobday, 1995). Second, this accumulation is the result of the interactions between individuals, organizations and firms, i.e. it is an interactive process drawing on internal and often also external competencies. Capabilities are generated by complex interactions within firms, technological collaboration arrangements between competing and complementary firms, and through linkages with innovation-oriented organizations such as universities and research centres (Kline and Rosenberg, 1986; Bell and Pavitt, 1993). Collaboration can be formal or informal (Storper and Venables 2004, Asheim et al., 2007) Third, firms tend to move along particular trajectories within which past learning contributes to particular directions of technical change (i.e. technology acquisition is path dependent) and the experience derived from those paths of change reinforces the existing stock of knowledge and expertise (Bell and Pavitt, 1993), hence leap-frogging is rare. In sum, innovation systems literature highlights the systemic aspects of innovation, its point of departure being that firms do not innovate in isolation and that the continuous interaction with the other components of the system often appears to be crucial. RIS scholars emphasize that interactions take place at local level. Tacit knowledge is better

6

shared among actors that share certain values, norms and, in general, the institutional framework. However, proximity is not enough yet physical proximity is seen as central for both sharing of tacit knowledge. The ability of any given firm to benefit from the knowledge available in the system is a function of its absorptive capacity, that is, its own technological capabilities. Assessing the technological capabilities of a region is a function of its components, its capabilities and its interactions, and these might differ significantly between developed and developing countries. It is the interplay between the regional technological capabilities and the firm technological capabilities that determines the capacity of the firm to benefit from the knowledge located in the region. As such, any method assessing the role of the RIS supporting the acquisition and development of technological capabilities in firms has to bring together both the firm and the regional technological capabilities. 2. 2. Adapting RIS to developing countries As discussed earlier, most RIS in developing countries do not show the high degree of systemicness that characterises RIS in developed countries. The technological level of the different organizations in the system is frequently low, their interactions are weak and they are, in general, more dependent on external flows of knowledge and technology. In this context, most of the assumptions in the literature need to be adapted to the specificities of developing countries and regions (Chaminade and Vang, 2006; Vang and Chaminade, 2007; Vang et al forthcoming). We will now turn to the most central dimension of RIS, synthesizing the general RIS-literature with special attention to the new attempts at adapting RIS to the specificities of developing countries and regions. By doing so, we will develop a set of hypotheses on the role of RIS in supporting upgrading of firms in developing countries. For each component of the RIS, we deduct one hypothesis derived from the existing literature. The hypotheses will be further tested in two regions in Mexico. 2.2.1. Systemicness in innovation systems in developing countries Much research within economic growth, upgrading, innovation and economic development has focused on either the supply or the demand side of the development process. In contrast, the RIS approach puts the emphasis on the systemic dimension of the innovation process (Lundvall, 1992, Asheim and Gertler, 2005); that is, the dynamic interaction between the different components in the system and the impact of the system’s strong or weak components on the dynamic efficiency of the system as a whole. Innovation systems research (Freeman, 1987, Lundvall, 1992, Nelson, 1993, Edquist, 1997) emerged as a response to the more linear model of innovation dominant mainly in the US until the eighties. IS research emphasized that innovation could occur outside the ‘labs or domain of science and technology; innovation systems research has especially stressed the interface between users and producers. Lundvall’s seminal text on user-producer interaction in the Danish dairy sector is one of the cornerstones in this literature (Lundvall, 1988).

7

Scholars within the RIS approach have mainly focused on the localized nature of these interactions. Emphasis has been especially on the tacit component of knowledge. Knowledge is considered to be embedded in specific institutional settings where local recipients share values and visions and organisational forms etc. that allow them to ‘decode’ the tacit knowledge available to them and thus increase their ability to tap onto tacit knowledge (Gertler 2004). Thus, most RIS researchers argue that interactive learning is facilitated by physical proximityvi. Well-functioning RIS such as the ones found in the developed world, with intensive interactions between the different organizations in the system are far from common in the developing context. In this sense, as we have argued elsewhere, RIS in developing countries should be understood as “immature RIS” or emerging RIS where some of the building blocks of the RIS are in place where the interactions among the elements of the RIS are still in formation and thus appear fragmented (Chaminade and Vang 2006a, Galli and Teubal 1997) thus fails to perform on the same level as mature RIS’s.

H1: There is a direct relationship between firms advanced capabilities and well-functioning RIS (i.e. we expect firms in RIS displaying a high degree of systemicness to have more advanced capabilities).

2.2.2. TNCs and the RIS Innovation studies have tended to emphasize endogenous growth dynamics focusing mainly on indigenous capacity building. However, several of the clusters that served as inspiration for RIS theoretical development are restructuring and reconstructing the boundaries between the local and the global. Well-functioning RIS such as Silicon Valley are increasingly been knitted with other global hubs such as Hsinchu Science Park in Taiwan and Bangalore in India (Saxenian 2006). The so-called global-local linkages have been elevated to the forefront of RIS studies, and this is considered especially critical for developing countries. As argued before, developing countries often lack local resources needed for acquiring advanced technological capabilities. They are much more dependent on external sources of knowledge. Apart from a yet immature and emerging line of research on globally distributed knowledge-bases, most of the literature has mainly focused on TNCs and spillovers to the local industry (Cantwell and Piscitello 2002).vii Within this latter stream of literature two different positions can be found. The more critical literature argues that while FDI and indigenous firms’ commercial interaction with TNCs can facilitate upgrading the upgrading is limited to technical upgrading as the TNCs – or lead firms – maintain control over their core competencies. Quantitative empirical studies tend to provide some empirical support for this. However the datasets used for the analyses are often so infused with problematic data that conclusive evidence has not yet been delivered (Zanfei 2005). Case studies and countries studies provide some anecdotal evidence that the opposite also exists, that is, that the collaboration with TNCs can lead to the development of core competences (initially OEM, then ODM) that, at the end, might be translated into indigenous branded products (OBP). The silicon chips production and Acer own brand strategy in Taiwan (Saxenian 2006) are examples of the latter stage. In the India’s IT service industry, it is possible to find several examples of

8

the importance of the latter form of collaboration and knowledge transfer for upgrading (Vang and Chaminade, 2006; Chaminade and Vang 2006b, Dossani and Kenney 2006). See also KJ Joseph’s chapter in this volume)). It is argued that the ability of developing countries to tap into, absorb and leverage global flows of traded and untraded knowledge and attract foreign direct investment (FDI) is one of the most important determinants of the performance of their upgrading. Yet, not all global interactions lead to the expected positive results. FDI’s, for example, are not a priori assumed to lead to positive direct or indirect spillovers as their impact will depend, among other issues, on the subsidiaries local embeddedness, the R&D-mandate, the decision-making structure of the TNC or, more generally, industry, institutional, temporal and firm specific characteristics. Based on this the following hypothesis’ can be deduced:

H2. The interaction between foreign subsidiaries and locally owned firms is important to acquire advanced technological capabilities in RIS in developing countries, yet it is not an automatic process

2.2.3. Users in innovation systems in developing countries Innovation systems research has long emphasized the importance of user-producer interaction for upgrading and innovation (Castellacci, 2006; Fagerberg, 2004; Lundvall, 1988; Jeppesen and Molin, 2003; Luthje et al, 2005, Thomke and Von Hippel 2002)). The emphasis of the user-producer interaction stems from the fact that innovations often occur in response to specific problems that emerge from the interaction between the user and the producer. This represents the foundation for breaking away from the linear innovation model and supply or demand models in general. Recently, the literature focus has shifted towards lead users (Franke and von Hippel 2003; Franke and Shan 2003; Franke et al 2005). Lead users defined as users that perceive needs well ahead of the mass market and that, often, have developed their own innovative adaptive solutions (Jeppesen and Frederiksen 2006). The interaction with users might support incremental innovations while interaction with lead users might be more important for more radical innovations and thus more valuable for the innovative firm (nevertheless most studies confirm that lead users are also mostly involved in creating incremental innovations (Jeppesen and Frederiksen, 2006). The user-producer model relies on the assumption that the user and the producer have ‘equal’ incentives for sharing the knowledge required for successful collaboration and that both have sufficient in-house human capital to absorb and use the exchanged information and knowledge or at least that the interaction constitutes a win-win situation. This approach has spurred an interesting and also critical debate concerning many different issues – for example on the relevancy of lead users’ preferences versus the mass markets preferences as well as studies of the importance of users in an evolutionary perspective (Chaminade and Vang 2006a). Given that export can be seen as a proxy for interaction with users at distant locations we propose that:

9

H3: Export to the world market is stimulating upgrading in firms located in RIS in developing countries (as more advanced users are located overseas).

2.2.4. Universities and innovation systems in developing countries Universities have always been considered a crucial element in innovation systems. These organisations play a major role in originating and promoting the diffusion of knowledge and technologies that contribute to industrial innovations (Mansfield and Lee, 1996, p. 1047). In particular, research universities are important as sources of fundamental knowledge and industry relevant technology in modern knowledge-based economies (Mowery and Sampat, 2004). Basically, universities might play three possible roles in a system of innovation: the generative, developmental and the evolutionary role. The generative role stresses the role of universities as producers of advanced basic research and trained personnel (i.e. graduates mainly)(Mowery and Sampat, 2005). It is not the objective of the generative role to supply the industry with knowledge solutions (in the sense of applied knowledge) but to produce science (basic knowledge) and to educate graduates. The generative role has been dominating the innovation system research, particularly among the early proponents, who tend to treat universities as a more or less autonomous system adhering to norms of academic research and teaching. Their research supports the claim that university research plays a small role in industrial R&D projects (Cohen et al, 2002; Fagerberg, 2004). Lundvall (2002) even goes so far to state that university’s “most significant contribution to society and the economy will remain well-educated graduates with critical minds and good learning skills” (2002, p…) (Vang et al, forthcoming). It follows that the so called Third Task is not considered to be relevant for the majority of industries (See chapter XXX in this Handbook for a more detailed discussion of the role of universities in systems of innovation in developing countries). In contrast, the developmental perspective puts a stronger emphasis on the impact of the university’s in the governance of the regional innovation system and in the close interaction between university and industry in the development of industry specific knowledge. The developmental role is the flagship of the triple helix model. The triple helix stresses the interaction between university-industry-government where knowledge production and innovation transcends organizational boundaries (Etzkowitz and Leydesdorff, 2000). While the generative role treats universities as more or less independent units, the triple helix model emphasizes the emergence of hybrid, recursive and cross-institutional relations between university-industry-government. In other words, the institutional boundaries (in terms of knowledge production and innovation) between university and industry are blurring, and the roles that both organizations take are no longer clearly identifiable. The triple helix model seeks to spur innovation of the more radical kind by conceptualizing the university as an incubator or seedbed that provides support structures for overlapping networks of academic research groups and start-up firms. Yet, very little has been reported on the disadvantages and conflicts of interests

10

related to academic entrepreneurialism (Gunasekara, 2006) or on the adequacy of the model to the specific territorial and historical context of the RIS. The latter issue is particularly relevant for developing countries. As recent research has pointed out (Saad, 2004 and Juma et al, 2001 cf Sardana and Krishna, 2006; Turpin and Martinez-Fernandez, 2003) the use of the triple helix concept in a developing context is problematic. In most of the cases, the interaction between the government, the university and the industry does not materialize due to the lack of resources, power and the weaknesses of the different actors involved in the system. Hitherto, the discussion on the generative versus developmental role of the university has been rather abstract and not taking into account the specific socio economic and historical context in which the interaction between universities and the other organizations in the system take place (Vang et al, forthcoming). In this vein, we propose an evolutionary model where the role of the universities evolves over time and as a response of the needs of the firms located in the RIS (Vang et al forthcoming). In the early phases of the emergence of the RIS, universities might play a crucial role as providers of qualified human capital (generative role). However, as firms acquire more advance technological capabilities and move up to more innovation intensive activities, they might require from the universities more industry specific research, thus, pointing out to the importance of a more developmental role. Overall, the situation in developing countries is one of a fragmented system of innovation, where in most cases, it is possible to identify a handful of firms with advanced technological capabilities and for which the developmental role of the university is of crucial importance. On the other hand, most firms in RIS in developing countries have basic or intermediate technological capabilities and require from the universities a much basic generative role (provision of qualified human capital) (Vang et al, forthcoming). Thus H4. Universities are not expected to play a fundamental role in the development of technological capabilities in firms located in RIS in developing countries

2.2.5. State intervention in innovation systems in developing countries Contrary to other system approaches such as Luhmann’s (1995) which implies self-regulating and closed systems, innovation systems research postulates that systems cannot be seen in isolation from their institutional framework, thus the idea of self-organizing systems is considered as rather meaningless.viii Traditionally, innovation system research has highlighted the role of policies targeting systemic problems (Chaminade and Edquist, 2006). While the NSI approach emphasizes the role of the national state (i.e. central government bodies) and devotes much attention to defending and rethinking the role of the national state in the context of increased globalization (Archibugi and Lundvall, 2001, Lundvall and Borras, 1999) RIS emphasizes the importance of regional authorities in constructing and supporting systems at a local level (Asheim et al, 2003)ix. The role of the state in regional systems of innovation has been extensively discussed particularly in the so-called ‘Italian district literature’. While there are different positions within this literature - Becattini’s only pays scant attention to the

11

state, while Bagnasco (1988), Brusco (1982) and Trigilia (1990) in particular have written extensively about the state – most underline the centrality of the local state (not the national state) in supporting interactive learning and facilitating innovation and how it comes to represent local interests as a whole, mediating between small-entrepreneurs, workers, and artisan interests and thus functions as means for correcting systemic errors. Uniting most RIS researchers is a disbelieve in the efficiency of markets as mediating the transactions that are conducive to innovation. In a detailed investigation of the majority of Asian countries Lundvall et al (2006) concur and find that the state cannot a priory be attributed a developmental role. Yet, Lundvall et al (2006) also find that in nearly all the cases of successful development in Asia the states have played a central role, particularly regional governments have shown to be central actors in the development of RISx. Thus the following contrasting hypothesis can be deduced:

H5: Regional innovation policy or initiatives (i.e. state intervention) are central elements for upgrading firm’s technological capabilities

3. Assessing technological capabilities in firms and Regional Systems of Innovation: a new method 3.1. Developing the method This section aims to provide a methodological framework to assess systematically the technological capabilities of firms and regional systems of innovation. It draws on the literature on systems of innovation and technological capabilities to develop a new method that integrates micro- and meso-level factors. Although private firms constitute the main component of regions technological capabilities, at the meso-level many other types of actors interact with each other within a specific socio-economic and institutional framework: universities, public research centers, government, industry associations, among others, as we have discussed earlier. Depending on the aims of the research, it is possible to emphasize the role of one component, but a meso-level analysis implies a systemic approach. For example, in FDI-led, high-technology manufacturing industries in less advanced countries, TNCs might be critical to the creation of technological capabilities. Their interactions with and indirect impact on the other components in the regional system are crucial. TNCs might have an effect on host economies through a wide array of formal and informal mechanisms such as technical assistance to local companies, knowledge and skills acquisition by local personnel working for the TNCs and imitation of new technologies by locally-owned firms.xi But the role of other components and the interactions among them are also central to the study of RTCs, given their systemic nature. We might expect that there are some important learning processes that are external to the firm and have to do with its relationships with other components in the system. Even large TNCs need to interact with and tap into resources from the local economy. In addition, absorption, adaptations, improvements and retention of foreign technology are not automatic and costless processes. Domestic firms and innovation-oriented organizations must engage in

12

deliberate and integrated efforts and devote substantial resources to start up and sustain a gradual process of knowledge accumulation, conducive to indigenous capability building (see among others, Young et al., 1994; Hobday, 1995; O’Donnell and Blumentritt, 1999; Narula, 2001). Table 1 presents a taxonomy to assess regional systems based on their technological capabilities.xii The columns list the main components of the system, while the rows describe the capability level – advanced, intermediate and basic – for each component. The capability level of each component is given by its relationships with other components, and the attributes of both components and relationships. The basic level portrays a region with, technologically speaking, weak actors while the advanced level describes a mature regional innovation system in terms of both relationships and attributes. The matrix does not claim to define the optimal role of each component, but rather to identify different levels of capabilities. This tool is useful insofar as it facilitates a structured and systematic comparison between regionsxiii. Information consists of original data collected in two Mexican regions - Jalisco and Baja California - in the autumn 2004 through of a survey. The survey inquires into their level of technological capabilities as well as their interactions with the other components of the regional innovation system. This provides the input for columns 1 and 2 of the table and for the quantitative analysis discussed in this chapter. The level of technological capabilities of the other organizations in the system is collected through semi-structured interviews with key personnel of the other regional actors as well as the analysis of existing statistics and secondary literature. The first two columns in Table 1 display the two main components of the regional innovation system: MNEs and locally-owned firms. Firm technological capabilities are assessed using firm-level information according to Table 2. Firm-level technological capabilities involve knowledge and skills both codified and tacit, and there is no single variable that summarizes and captures their complex nature. Based on the distinction between capabilities and competencesxiv, outcome-related variables, such as the introduction of new products or improvements to existing equipment, are used to evaluate technological capabilities. Two types of technological capabilities are distinguished: a) process and production-centred and b) product-centredxv. The latter relate to the knowledge and skills needed to produce existing goods and to carry out technological product innovations. In turn, process and production organisation capabilities are the knowledge and skills needed to operate production processes efficiently and to create new or significantly improved processes. They comprise the knowledge needed to use, improve or innovate machinery and equipment on the one hand, and to implement, modify and create new methods of production organisation on the other. The use of advanced management techniques is included here within process and production organisation capabilities. Firm-level capabilities are also classified into three levels – basic, intermediate and advanced – according to their technological complexity.xvi This classification aims to differentiate between production capabilities (to produce goods using existing

13

technologies) and innovation capabilities (to generate and manage technical change). It follows that there will be industrial differences in the specific capabilities to consider in each level. The taxonomy presented here has been customised for sectors such as the electronics industry, characterised by great flexibility to decompose the value chain across national borders, high R&D expenditures and widespread use of complex production organisation techniques.xvii

At firm-level, the questionnaire collects information both on the level of technological capabilities in the firm and on the determinants of technological capabilities (internal and external). The potential factors associated with technological capabilities are summarised in Table 3. It does not claim to be an exhaustive list but on the basis of the existing literaturexviii - and taking into account the characteristics of the phenomenon studied - the most important are included. The factors were divided into two: internal and external to the firm, and included all factors related to the hypotheses presented above (interactions with local organizations, government support, exports to the world market, etc.) Table 3 lists and describes these factors. The third column in Table 1 deals with Universities and Technical Education Centres and their interaction with the industry. For the purpose of the methodology that we are proposing here, it is important to remember that this research focuses only on those departments or units, within each component, directly related to the studied sector. For instance, when a university or technical education school is analysed, it focuses on the engineering departments and units directly related to the studied sector. The fourth column in Table 1 presents the attributes and relationships among public research centres. R&D activities can be conducted in research universities, research laboratories in private firms or public research laboratories. Research centres conduct diverse activities – such as basic and applied research, development of prototypes, formation of highly-qualified human resources through teaching, and development of new instruments and techniques, and have a substantial impact on industrial R&D in technology-intensive industries such electronics (Cohen et al., 2002). The fifth column in the table refers to the public sector. As discussed in the previous section, national and local governments play quite different roles in the development of technological capabilities. On the one hand, the public sector is responsible for creating and supervising institutions that foster technological capabilities, such as S&T law, protection of IPR, competition law, a research council or ministry of S&T, etc. On the other hand, governments can promote the use, diffusion, improvement and production of scientific and technological knowledge through science, technology and innovation policies.xix The qualitative and quantitative indicators used to assess the public sector must take into account that this research studies regional capabilities in developing countries, where the features of institutions and policies are different from those in developed countries.

Table 1 Regional technological capabilities

Components/

Level Foreign subsidiaries Local firms Universities and

technical education centres

Public research centres Public sector Private organisations

Advanced - Advanced technological capabilities within foreign subsidiaries - Strong backward linkages and integration with the local economy - Abundant knowledge flows from foreign subsidiaries to the other components of the regional system (both research- and teaching-oriented) - Complementarity and strong linkages with local research (public and private research centres, research universities) - Strong inter-firm knowledge linkages with other foreign subsidiaries and locally-owned firms

- Advanced technological capabilities within local firms - Local firms design and manufacture final goods and components to be sold in the local market and abroad - Strong research-oriented linkages with other components of the system - Joint collaboration with foreign subsidiaries in design and product development - Strong trade and knowledge linkages with other locally-owned firms (local networks)

- Large number of universities and technical education schools offering highly-qualified and specialised scientists, engineers and technicians (university degrees and postgraduate programmes) - Rapid response to changes in technologies, and even anticipation of those changes - Strong basic and applied research activities - Strong research- and teaching-oriented linkages with firms, including collaborative research projects - Frequent involvement in technical assistance projects with industry

- Several sector-oriented public research centres - Formation of highly-qualified and specialised resources for the sector (DPhil and master’s) - Abundant collaborative projects with industry - Commercialisation of outputs (licences, patents, instruments, etc.) - Focus on basic and applied research, and significant presence of commercial oriented activities - Frequent involvement in technical assistance projects with industry - Important number of researchers leave the centre to establish their own company (indirect spin-offs)

- Strong S&T institutions and public offices at the regional level - Strong planning, designing and implementing of innovation-oriented initiatives - Strong coordination among public offices in charge of implementing innovation-oriented initiatives - Strong support to develop highly-qualified and specialised human resources - Active science, technology and innovation policies properly customised to meet the needs of the region and the sector

- Sectoral industry associations with strong presence in the region - Industry associations and other private organisations provide strong support to technological capability building - A strong group of local managers which promotes technological capability building in the region (within foreign subsidiaries, in locally-owned firms, universities, research centres) - Frequent direct participation of foreign subsidiaries personnel in regional initiatives to strengthen capabilities in local firms - Strong and abundant capital suppliers to fund innovation projects, spin-offs or start-ups.

Intermediate - Intermediate technological capabilities within MNEs - Some backward linkages with the local economy

- Intermediate technological capabilities within local firms - Local firms manufacture or assemble components

- Good number of universities and technical education schools offering scientists, engineers and technicians with general knowledge

- A few sector-oriented research centres carrying out basic and applied research which is relevant for the industry established in the region - Collaborative research

- Some S&T institutions and public offices at the regional level - Planning and designing of regional science, technology and innovation policies - Some of the initiatives are

- Sectoral industry associations with strong presence in the region - Industry associations and other private organisations provide some support to technological capability

- Teaching-related links with universities and technical education centres - Few collaborative projects with universities and research centres - Some inter-firm knowledge linkages with other foreign subsidiaries and locally-owned firms

mainly for foreign subsidiaries located in the region or other regions within the country - Some linkages with universities and research centres, but mainly teaching-oriented - Strong flows of technology from foreign subsidiaries to local firms - Weak local trade and knowledge networks

- Not enough specialised highly-qualified personnel - Slow response to changes in technologies (to adjust programmes and courses) - Some basic and applied research - Strong teaching-oriented links and some research-oriented links with firms

projects with industry, mainly in response to the needs of firms - Formation of highly-qualified and specialised resources for the sector (DPhil and master’s)

not implemented because of lack of resources - Reduced budget and resources to promote innovation in the sector

building - A group of local managers which promotes the development of the industry, working mainly in areas not directly related to innovation: infrastructure, public services, regulation, etc. - Weak and few capital suppliers to fund innovation projects, spin-offs or start-ups.

Basic - Basic technological capabilities within MNEs - Poor backward linkages with the local economy (enclaves) - Limited knowledge flows from MNEs to the other components of the regional system

- Basic technological capabilities within local firms - Very few local companies supplying services and indirect goods to foreign subsidiaries - Weak or non-existent links with the rest of the system - Limited flows of technology from foreign subsidiaries to local firms

- Few universities and technical education schools - Lack of sectoral specialisation - Weak or non-existent sector-oriented research - Limited teaching-oriented links with industry, and lack of research-oriented linkages

- Few, or even lack of, public research centres - Weak or non-existent linkages with industry - Strongly focused on basic research without commercial applications

- Weak, or even lack of, regional S&T institutions or public offices; weak or non-existent coordination among public offices - Very few, or even lack of, science, technology and innovation policies to meet the needs of the region and sector - Limited or non-existent budgets to promote innovation in the sector - Poor involvement of industry, private organisations and academia in the formulation of public policies

- Sectoral industry associations with weak presence in the region - Industry associations are mainly oriented to provide legal or administrative advice (few or non-existent activities to promote innovation in the sector) - Weak coordination among the sectoral private organisations - Lack of capital suppliers to fund innovation projects, spin-offs or start-ups.

15

Source: Padilla-Perez (2006).

Table 2

Firm-level technological capabilities Types of capability

Levels of capability Process and production organization Product-centered

Basic

- Sub-assembly and assembly of components

and final goods - Minor changes to process technology to adapt

it to the local conditions - Maintenance of machinery and equipment - Production planning and control - Efficiency improvement from experience in

existing tasks

- Replication of fixed

specifications and designs - Minor adaptations to product

technology driven by market needs

- Routine quality control to maintain standards and specifications

Intermediate

- Manufacture of components - Improvement to layout - International certifications (ISO 9000) - Introduction of modern production

organizational techniques (e.g. just in time, total quality control, etc.)

- Automation of processes - Flexible and multi-skilled production - Selection of technology (capital goods)

- Product design department

(design for manufacturing) - Development of prototypes - Improvement of product quality

Advanced

- Own-design manufacturing - Major improvements to machinery - Development of equipment - Development of new production processes - Development of embedded software - Radical innovation in organization - Process-oriented R&D

- Development of new products

or components - R&D into new product

generations - Research into new materials and

new specifications

Source: Padilla-Perez (2006), based on Lall (1992), Bell and Pavitt (1995), and Ariffin and Figueiredo (2003).

Table 3 Potential factors associated with technological capabilities at firm level

Internal to the firm

Variable Definition Age Age of the plant since it was established in Mexico: 2004 minus year in which

the firm was established. Exports Percentage of total production exported. Growth Employment growth between 2002 and 2004. Human capital: - Direct/indirect - Unqualified/qualified

Two indicators to measure human capital: - Direct over indirect employees: (blue collar workers) / (supervisors + technicians + engineers + administrative personnel).

17

- Unqualified personnel over highly qualified personnel: (technical education + high school + primary school + no education) / (postgraduate degree + university degree).

Ownership A binary variable that takes the value 0 if the firm is foreign-owned and 1 if it is locally-owned.

Size Number of employees in 2004. Training expenditures Average number of hours per employees per year.

External to the firm Source universities A binary variable that takes the value 1 if the plant has used universities as a

source of technology and 0 otherwise. Source research centre A binary variable that takes the value 1 if the plant has used research centres as a

source of technology and 0 otherwise. Number of sources A variable summarizing the total number of external sources of technology used

by the firm. It corresponds to the simple sum of sources, and has a maximum value of 11 and minimum of 0. The sources of technology are: suppliers of equipment and inputs, public research centres, universities, recruitment of highly-qualified personnel, licensing, clients, competitors, consultancies, fairs and exhibitions, industry associations, and other.

No. links universities A variable summarizing the total number of different links that the firm has with local universities. It represents the simple sum of links, and has a maximum value of 5 and minimum of 0. The links include: training, student internships, secondment or visiting programs for professors, collaborative research projects, and other.

No. public initiatives A variable summarizing the total number of different public initiatives to foster innovation or technology dissemination in which the firm has participated. It corresponds to the simple sum of initiatives, and has a maximum value of 6 and minimum of 0. The public initiatives are: training, tax incentives, funds to develop new products, technology diffusion, technology upgrading, and other.

Technology transfer A variable summarizing the total number of different types of technical assistance that the foreign subsidiary has offered to its local suppliers and the total number of different types of technical assistance that a locally-owned firms has received from TNCs established in the region. The different areas of technical assistance considered are: product specifications, quality control, process and production organization, training of engineers and technicians, purchase of machinery and equipment, and procurement of components and raw materials. It corresponds to the simple sum of the different types of technical assistance, and has a maximum value of 6 and minimum of 0.

Region A binary variable that takes the value 0 if the firm operates in Jalisco and 1 if it is located in Baja California.

The last column in Table 1 refers to industry associations and other private organisations that underpin the innovative strategy of private enterprises. These organisations may provide several types of services, such as training; diffusion of technology; services of normalisation, certification and standardisation; technical assistance for technological upgrading; promotion of a culture of quality, etc. For small enterprises in developing countries, initiatives to assist the process of international certification and training of human resources are very important. Regarding their relationships with other components of the regional system, industry associations may, for instance, foster university-industry

18

links, assist private firms in the application and administrative processes involved in getting public support, collaborate with the government in designing and implementing initiatives for the sector, etc. These organisations may act as bridges between users and producers of knowledge, and are commonly known as bridging institutions.xx

Capital suppliers are included within this group of private organisations. It is crucial for a system of innovation to possess a financial system that has the resources and willingness to finance innovation. Innovation is an expensive process and significant resources must be devoted to initiate, direct and sustain it. It is also a long-term and slow process (and the resources for its support must be committed over a similarly long term) and its outcomes are uncertain (O’Sullivan, 2005, p. 240). Large firms finance internally risky investment in innovation, but small firms, especially in developing countries, do not have the financial resources to do this (Christensen, 1992). Obtaining credit is generally a problem for firms in developing countries and finding the resources for incremental innovations is probably the most difficult. Such activities are often considered too risky, intangible and as requiring long-term finance (Luthria and Nabi, 2002). 3.2. Analysing technological capabilities in regional systems of innovation. Testing the hypothesis 3.2.1. General view of the regional innovation system Information collected this way allows the researcher to classify the different components of the system according to a scale from basic to advanced capabilities. The methodology was applied to two Mexican regions, Jalisco and Baja California in Autumn 2004. The comparison of the results for the two regions is depicted next.

Graph 1 Regional technological capabilities

Foreign subsidiaries Local firms

Universities & technical

schools

Research centres

Public sector

Industry associations

Components

Level

Advanced

Intermediate

Basic

Jalisco Baja California

Foreign subsidiaries Local firms

Universities & technical

schools

Research centres

Public sector

Industry associations

Components

Level

Advanced

Intermediate

Basic

Jalisco Baja California

19

Source: Padilla-Pérez (2006)

The information to assess the first two components comes from the firm survey applied to 80 firms located in the studied regions. Additionally, 30 semi-structured interviews were conducted with key innovation actors in the system of innovation. The firm questionnaires aimed to collect two main types of firm-level information: indicators related to technological capabilities and factors potentially associated with technological capabilities. As a first step, it was necessary to identify the relevant population, since there was no list that comprises all the electronics firms in each state.xxi Two criteria were used to classify firms in order to have a representative sample: type of firm and origin of capital.xxii The 36 firms interviewed in Jalisco (of which 55% were foreign-owned) represented 82% of the relevant population and had altogether 26,993 employees at the end of 2004. In Baja California the sample included 44 firms (72% foreign-owned), representing 24% of the population and with an overall employment of 40,621. In Jalisco, both foreign subsidiaries and locally-owned firms had higher technological capabilities than those in Baja California. In Jalisco, 45% of interviewed firms had basic product-centred capabilities, 17% intermediate and 38% advanced, while 23% had basic process and production organisation capabilities, 52% intermediate and 25% advanced (see Table 7). Only 4% of interviewed firms in Baja California had advanced product-centred capabilities, while 75% of them had basic capabilities. On the other hand, 27% of interviewed firms had basic process capabilities, 61% intermediate and 11% advanced.

Table 4 Firm-level technological capabilities in Jalisco and Baja California

Jalisco Baja California Product-

centred Process Product-

centred Process

Advanced 38% 25% Advanced 4% 11% Intermediate 17% 52% Intermediate 21% 61% Basic 45% 23% Basic 75% 27% 3.2.2. Testing the hypothesis To test the hypotheses, two complementary analyses were made. First an econometric analysis of the main factors associated with technological capabilities at firm level as well as the linkages of the firm with the other organizations of the system, collected through the survey. Second, the econometric analysis was complemented by information collected from semi-structured interviews with other actors in the regional system of innovation. As for the econometric analysis, the following model was proposed: TCi = β0 + β1 FA1i + … + βn FAni + α1 Ry + εi ;

20

where TCi is an index of technological capabilities in firm i; FAxi are firm-specific factors associated with technological capabilities (the number of factors ranges from 1 to n); Ry identifies the region in which the firm is established and is a binary variable since the fieldwork collected empirical evidence on two regions; and εi is the error term. The technological capability index compares capabilities across firms using systematic criteria to classify or rank them. Its categories can be ranked from low to high, but the distances between adjacent categories are unknown, i.e. the index comprises relative values. Consequently, it is argued that the index should be treated as an ordinal variable.xxiii Table 5 summarises the results for the whole sample (i.e. the 80 interviewed firms). The interpretation of the results will be done for each hypothesis.

Table 5 Factors associated with technological capabilities - The sample

Dependent variable

Coefficients (standard errors in brackets) Variables TC Overall TC Process TC Product

Age 0.003 (0.018)

0.022 (0.341)

0.004 (0.043)

Exports -0.014*** (0.005)

0.016* (0.010)

-0.054*** (0.013)

Growth -0.283** (0.136)

-0.885 (0.275)

-0.390 (0.351)

No. links universities 0.139 (0.166)

-0.020 (0.329)

0.385 (0.383)

No. public initiatives -0.027 (0.154)

-0.885*** (0.323)

0.546 (0.472)

Region (=Jalisco) 0.124 (0.366)

0.947 (0.746)

-1.095 (0.676)

Size 0.368** (0.155)

1.554*** (0.398)

-0.265 (0.362)

Source research centres (=No)

-0.865** (0.371)

- -2.229** (0.899)

Training expenditure 0.138 (0.138)

-0.125 (0.262)

0.325 (0.321)

Unqualified/qualified -0.025** (0.012)

-0.044* (0.025)

-

Number of sources

- 0.351** (0.163)

-

Direct/indirect

- - -0.288** (0.145)

Model fitting information -2LL intercept only: 226.41 -2LL final: 173.54 Significance: .000 Goodness of fit measure Pseudo R2 (Nagelkerke): 0.511 Parallel regression assumption met at 0.079 Ordinal probit regression

Model fitting information -2LL intercept only: 160.19 -2LL final: 118.27 Significance: .000 Goodness of fit measure Pseudo R2 (Nagelkerke): 0.469 Parallel regression assumption met at 0.386 Ordinal logit regression

Model fitting information -2LL intercept only: 150.73 -2LL final: 86.96 Significance: .000 Goodness of fit measure Pseudo R2 (Nagelkerke): 0.645 Parallel regression assumption met at 0.999 Ordinal logit regression

Note: * Significant at the 0.10 level, ** Significant at the 0.05 level, *** Significant at the 0.01 level.

21

Goodness of fit for this cross-sectional model and sample size was good. The independent variables explain 51.1% of the variation in overall technological capabilities, 46.9% of process capabilities and 64.5% of product capabilities. The difference between -2LL intercept and -2LL final was always significant at the 0.01 level. All regressions met the parallel regression assumption. H1. There is a direct relationship between firms advanced capabilities and well-functioning RIS (i.e we expect firms in RIS displaying a high degree of systemicness to have more advanced capabilities). The econometric results presented in Table 5 lead us to accept hypothesis 1. First, firms that use external sources of knowledge (such as research centres, clients, suppliers) have in average higher process technological capabilities. Second, firms that interact with research centres and universities in the studied regions in Mexico have in average higher product-centred capabilities. When considering the region of origin of the firm, the percentage of positive answers for all potential sources (suppliers of equipment and inputs, public research centres, universities, recruitment of highly-qualified personnel, licensing, clients, competitors, consultancies, fairs and industry associations) was always higher for Jalisco than for Baja California, showing the stronger isolation, in technological terms, of firms in the latter region. The difference between the two regions was especially noticeable for universities: 55% of interviewed firms in Jalisco said they used universities as a source of technology, but only 11% of firms in Baja California said they did so (More about universities below, see Table 6). Links among firms in both regions were important but mainly related to the coordination of manufacturing activities and outsourcing.

Table 6

Sources of technology (Percentage of positive answers)

Source Jalisco Baja

California Suppliers of equipment and inputs 89 82 Public research centres 33 14 Universities 55 11 Recruitment of highly-qualified personnel 83 55 Licensing 19 9 Clients 81 59 Competitors 47 45 Consultancies 50 32 Fairs, exhibitions 53 41 Chambers of commerce and industry associations

44 31

22

The additional information collected through the semi-structured interviews with other regional actors also confirm the a higher degree of maturity of the RIS in Jalisco compared to Baja California. Industry associations and other private organizations in Jalisco played an active role in promoting the development of the electronics industry in the region. There were two main sectoral associations which together grouped around 90% of firms active in the electronics industry (CANIETI Occidente and CADELEC).xxiv These associations provided a wide array of services to their affiliates in administrative, fiscal, local and federal regulations, as well as assistance to firms to strengthen technological capabilities and train human resources. In Baja California there were several industrial associations and private organizations. These organizations had launched initiatives to support manufacturing activities in the state, although none of them was exclusively oriented to assisting the electronics industry. They covered such diverse areas as market intelligence, education, attraction of foreign direct investment and economic development, but in general their initiatives had so far had a limited impact on the local industry. The additional interviews also provided some interesting information with regard to the importance of personal networks. The role of Mexican subsidiary managers and other managers in key positions within foreign subsidiaries in Jalisco is relevant to explain the differences in capabilities in the two regions. 86% of the interviewed foreign subsidiaries in this region were managed by a Mexican national. Mexican managers of foreign subsidiaries had had a crucial role in attracting new production lines and, more importantly, new technologies and higher value-added activities to the Mexican firm. Face-to-face interviews with subsidiary managers highlighted that subsidiary evolution, in terms of more technologically complex activities, had been a long and slow process. This process had been accomplished mostly by the activities of Mexican subsidiary managers in bargaining with and persuading parent companies of Mexico’s, and particularly Jalisco’s, capacities to take on and successfully perform new and more complex activities.xxv Subsidiary and other senior managers also participated actively in industry associations. Some of them met frequently with the objective of improving the competitiveness of the electronics industry in Jalisco. They had launched a series of coordinated actions in areas such as education and technology, infrastructure, and improvement of public regulation. In addition, there was a group of local managers who provided free technical support, through the local industry associations, to locally-owned companies producing electronic goods or supplying goods and services to the electronics industry. In sum, the systematic assessment of regional technological capabilities provides evidence to accept the hypotheses presented above. First, firms in RIS displaying a high degree of systemicness perform best (H1). A central factor that explains different firm performance (in terms of technological capabilities) in Baja California and Jalisco is stronger relationships (as well as the type of relationship) among the components in the latter. Firms not only interact with universities and research centres more frequently in Jalisco, but also research-oriented links (such as technical assistance and research collaborative projects) are more common. In the same line, firms in Jalisco carry out

23

coordinated actions – with other firms, academia and local government - in areas such as education and technology, infrastructure, and improvement of public regulation. H2. The interaction between foreign subsidiaries and locally owned firms is important to acquire advanced technological capabilities in RIS in developing countries yet it is not an automatic process To unpack the relationship between foreign subsidiaries and local firms the survey sample was divided by origin of capitalxxvi and new variables were included in the regressions: - Purchase local (only for TNC subsidiaries): A binary variable that takes the value 1 if

the foreign subsidiary has purchased products or services from local companies, and 0 otherwise (either direct or indirect goods).

- Previous experience (only for locally-owned firms): A binary variable that takes the value 1 if the owner or founder of the locally-owned firm had previous experience as an employee or supplier with TNCs before setting up his/her own firm, and 0 otherwise.

- Knowledge acquisition from TNC (only for locally-owned firms): A variable summarizing the total number of different types of knowledge that the owner or founder of the locally-owned firm acquired from his/her previous experience with TNCs, and he/she was currently using in his/her firm. It represents the simple sum of types of knowledge, and has a maximum value of 3 and minimum of 0. The different types of knowledge are: product-centered technology, process and organization production technology, and market knowledge.

Table 7

Factors associated with technological capabilities – Locally-owned firmsxxvii

Variables Dependent variable Coefficients (standard

errors in brackets) TC Product Exports -0.067 (0.022) *** Knowledge from TNC 1.513 (0.654) ** Number of sources 0.213 (0.326) Training expenditure 1.971 (0.732) *** Model fitting information -2LL intercept only: 55.64 -2LL final: 27.65 Significance: .000 Goodness of fit measure Pseudo R2 (Nagelkerke): 0.740 Parallel regression assumption met at 0.498 Ordinal logit regression

Note: * Significant at the 0.10 level, ** Significant at the 0.05 level, *** Significant at the 0.01 level.

24

The only variable significantly associated with advanced technological capabilities is knowledge acquired by local entrepreneurs through their previous experience with TNCs (see Table 7)xxviii. Two factors help to explain these results. 98% of TNCs interviewed purchase goods from locally-owned firms (mainly indirect goodsxxix), and almost all of them offer technical assistance to their suppliers. Thus, they transfer technology to local firms independently of their technological capabilities. Dummy variables for each type of technology transfer were introduced, but they were not significant. Second, locally-owned firms operating in the electronics industry receive technology from TNCs, but its type and complexity was relatively homogenous among firms interviewed and was not significantly associated with advanced capabilities. The qualitative analysis allows us to establish that interaction between TNCs, and locally-owned and local organisations is important to develop advanced technological capabilities (H2). The additional information collected through the interviews with other actors in the system also provides interesting information. The two regions studied are interesting case studies of two different types of global–local interactions and the related outcomes in terms of regional capability building. Almost 40 years after the first foreign subsidiary active in the electronics industry was established, Baja California has developed limited technological capabilities. Foreign subsidiaries in Baja California operate as enclaves: they import all, or almost all, of their inputs and intermediate products; forward and backward linkages with local firms are limited or non-existent; and links with local organisations such as universities and research centres are weak. As regards Jalisco, at the time of the fieldwork a significant production and technological transformation was taking place, through a virtuous circle between foreign subsidiaries and local agents. On the one hand, foreign subsidiaries had moved towards higher value-added activities and increased their interactions with local actors. On the other, the presence and activities of foreign subsidiaries have stimulated and supported the creation of better human resources and innovation-oriented organisations. By a process of cumulative causation, higher regional technological capabilities have encouraged foreign subsidiaries to transfer more technologically advanced activities to firms in the region H3. Export to the world market is stimulating upgrading in firms located in RIS in developing countries (as more advanced users are located there). As for the third hypothesis, Table 5 shows that the coefficient of exports is negative and significant for overall and product capabilities. The sign of the coefficient contradicts H3 and economic theory, which assert that exports, through access to new and bigger markets, generate economic incentives for increased innovative effort. Firms in Jalisco have in average higher product capabilities, but export a lower proportion of their production than firms in Baja California, which are more integrated into the US economy. The negative relation between exports and product capabilities is especially strong for small, knowledge-intensive firms in Jalisco, which are engaged in product design,

25

product development and R&D, but sell most of their products (or services) to MNEs established in the same or other regions within Mexico. In contrast, exports are significantly and positively associated with process capabilities. In general, process capabilities in the electronics industry are associated with large plantsxxx, which possess the financial and human resources to implement complex production organisation techniques and undertake long and costly certification processes. Large firms, which are mainly foreign subsidiaries, are more oriented to foreign markets, since they set up plants in Mexico to supply the US market. Summarising, exports stimulate process technology upgrading, but not necessarily product technology upgrading.xxxi

H4. Universities are not expected to play a fundamental role in the development of technological capabilities in firms located in RIS in developing countries As presented in Table 5, the coefficient of research centres as a source of technology is negative and significant for overall and product-centred capabilities. Source research centres and number of links universities are highly correlated, when the former is dropped from the regression, the latter is significant for product-centred capabilities. Firms with advanced capabilities, in particular product-centred capabilities, use research centres and universities as a source of technology. Advanced product-centred capabilities were less common among the interviewed firms, and on average these firms used research centres as the main source of knowledge. The interviews with representatives from research centres established in the two studied regions showed collaborative research projects involving research centres and firms were heavily concentrated on product-centred technologies. Process-related knowledge came from other sources of technology such as suppliers of machinery and equipment and consulting firms. Information collected through the additional interviews and the analysis of secondary data confirms that Jalisco has a large number of universities and technical education schools that provide firms active in the electronics industry with specialist and well-qualified personnel. According to an industry association (CADELEC) in 2005 there were 10 universities and 16 technical education schools in Jalisco offering programmes related to the electronics industry. These organisations had technical and undergraduate programmes clearly oriented to meeting the needs of industry and there were several postgraduate programmes related specifically to the electronics industry. All the interviewed organizations have strong interactions with industry, such as formal and informal agreements for student internships, short and customized training courses, or updating and modifying the content and structure of taught programs through the participation of private enterprises. In addition, the universities in Jalisco that were interviewed carried out applied research related to the electronics industry and two were also involved in basic research. These latter two had also been involved in collaborative research projects with industry. Table 8 summarises the results from the interviews with innovation-oriented organisations. Baja California had fewer universities and technical education schools offering programmes oriented to the electronics industry, despite its electronics industry being bigger than Jalisco’s.xxxii According to a local government report, in 2004 the state of

26

Baja California had six universities and 10 technical education schools offering programmes related to the electronics industry.xxxiii University engineering departments were oriented to educating students to work for manufacturing industry in general, but not with a specific focus on the electronics industry. There were a few postgraduate programmes available, but again they were not specific to the electronics industry. All the universities and technical education schools that were interviewed stated that they interacted with firms to update the content and structure of taught programmes via several formal and informal mechanisms. Three out of four of the universities that were interviewed in Baja California carried out applied research related to the electronics industry, but not necessarily to the activities of firms in the region; only one university did basic research. None of these universities had been involved in collaborative research projects with industry. The applied research that was conducted was mostly related to projects with educational objectives.

Table 8 University/technical education schools – industry links

(Percentage of positive answers)

Jalisco Baja California Curricula updating 100 100 Student internships 100 100 Donation of equipment 100 100 Training courses 100 100 Secondment programmes for professors

50 0

Basic research * 67 25 Applied research * 100 75 Collaborative research projects * 67 0 Technical assistance 50 60 Participation in public initiatives to promote interaction with industry

100 29

* The percentage of positive answers to basic research, applied research and collaborative research projects takes into account only universities, since technical education schools in the studied regions are supposed to be purely teaching oriented (according to the activities set out in their charter). Each region had two research centres specialised in or conducting research on areas related to the electronics industry, all of which were interviewed. In Baja California, both centers offer Master and PhD degrees and are heavily oriented to basic research. Their interactions with industry were limited and almost restricted to offering customized training courses. In clear contrast, public research centres in Jalisco carry out basic and applied research, and are involved in collaborative research projects and technical assistance with local firms (mostly foreign subsidiaries, but also with some locally-owned. Both were founded as the outcome of the interaction between TNC foreign subsidiaries and Mexican universities, and one of them (CINVESTAVxxxiv) offers postgraduate programmes in electronics. So, we can also conclude that firms located in RIS with strong presence of universities and public research centres perform better in terms of technological capabilities (H4 is

27

thus rejected); The interaction between firms and universities and research centres could lead to a virtuous circle of technological capability building. On the one hand, research laboratories within universities or public research centres act as a conduit for technologies from foreign subsidiaries. Joint research projects with firms provide research labs (in universities and public research centres) with financial resources and state-of-the-art technologies, which are crucial given their limitations to access both (resources and technologies). On the other hand, research labs offer high value-added services to local firms and foreign subsidiaries, anchoring the latter to the host region and assisting the former to develop their own technologies. Universities and research centres in Jalisco had been greatly benefited from the interaction with TNC foreign subsidiaries established in the region, and the latter had moved to more knowledge-intensive activities since they had found highly-qualified human capital and specialized organizations that supported their technology strategy. The interaction with the universities and research centers has a positive impact on product-centered capabilities but not so much for process centred. H5: Regional innovation policy or initiatives (i.e. state intervention) are central element for upgrading firm’s technological capabilities In order to obtain more robust results to test hypothesis 5 (the role of regional innovation policy), the variable number of public initiatives in Table 5 was disaggregated by type of initiative, that is a dummy variable was introduced to examine whether a particular initiative has a positive impact on firm-level technological capabilities. The five public initiatives were: training, tax incentives, funds to develop new products, technology diffusion and technology upgrading. The coefficient was significant only for government new products (use of public funds to develop new products) for product-centred capabilities. Table 9 summarises the results, only for the regression in which the coefficient was significant.

Table 9 Factors associated with technological capabilities – The impact of public policy

Variables Dependent variable Coefficients (standard

errors in brackets) TC Product Age 0.039 (0.047) Direct/indirect -0.989** (0.465) Exports -0.070*** (0.016) Government new products (=No) -3.449*** (1.318) Growth -0.927 (0.581) No. links universities 0.264 (0.400) Number of sources -0.272 (0.197) Region (=Jalisco) -1.571* (0.938) Size -0.448 (0.393) Source research centres (=No) -3.431** (1.097) Training expenditure 0.376 (0.370) Model fitting information -2LL intercept only: 147.76 -2LL final: 77.96 Significance: .000

28

Goodness of fit measure Pseudo R2 (Nagelkerke): 0.696 Parallel regression assumption met at 0.624 Ordinal logit regression

Note: * Significant at the 0.10 level, ** Significant at the 0.05 level, *** Significant at the 0.01 level. In the first set of regressions (Table 5), number of public initiatives was significant, but negative, only for process capabilities. Therefore, government support is significantly associated with advanced product-centred capabilities, but not with process capabilities. Moreover, among the initiatives to foster innovation, public funding to develop new products is the most important among the public initiatives. As for process capabilities, these are more homogenous in the sample, since the electronics industry operates under high international standards. Government support seems not important to explain the type of process technological capabilities possessed by firms interviewed for this research. As regards to the comparison of the two regions, the information collected from other regional actors revealed important differences between them. The local government in Jalisco had an office in charge of science and technology policy (State Science and Technology Council of Jalisco). This Council coordinated a series of public initiatives such as public funds for R&D, promotion of university-industry links, technology dissemination and human resource formation. It had a well-developed program to foster innovation in manufacturing, with specific initiatives for the electronics industry, although it should be acknowledged that the local government budget was limited and many initiatives in the programme described above had not been implemented due to lack of resources. Baja California also had a programme of science and technology policy, but it does not have any specific governmental agency that coordinates its implementation. Local policies to support innovation and formation of human resources were scant and spread across different local ministries, such as education and economic development. In addition, there was a shortage of public funds to support innovation. As a direct consequence, for firms in Baja California it is harder to find government support;, actually none of the firms interviewed had participated in innovation-oriented initiatives. So, regional innovation policy is a central element for firm technological capabilities in RIS in developing countries (H5). The comparison between Baja California and Jalisco shows that regional innovation policy is a central element to building technological capabilities. Strong institutions and an active local public sector are central to creating the framework for and fostering innovation among firms and organisations in the region. Some of the attributes present in Jalisco, but not in Baja California, are illustrative of the importance of institutions and an active public sector: a ministry that coordinates industrial public policy; a public office in charge of science, technology and innovation policy; public initiatives aimed at fostering innovation in the electronics industry including dissemination of technology, promotion of university links and technological upgrading, and formation of human resources, among others. An important lesson from the case of Jalisco is that abundant government financial resources are not a necessary condition to promote innovation and the development of technological capabilities. A solid and clear institutional framework, support for local initiatives and, more

29

importantly, the coordination of local efforts may be as or even more significant in terms of their effects. Co-funding of local initiatives, frequent communication with academia and the private sector, and the strengthening of links among organisations do not demand large sums of financial resources, but can be very effective mechanisms for strengthening regional capabilities. 4. Conclusions and policy implications The paper set out to investigate the relevance of and the need to adapt regional innovation systems to the specificities of the developing countries. This is inspired by the increased focus on regional innovation systems as a means for upgrading in developing countries by academics, consultants and policymakers. In line with other recent stream of research, RIS emphasizes the positive correlation between innovation and development. Yet, RIS’s specific trademark is the importance of the interaction between the local components of the system. RIS’s recent popularity has spurred a vast amount of anecdotic studies but only limited systematic comparative econometrical research. Based on a combination of the generic RIS research and the branches focusing more on constructing innovation systems in developing countries we identified the five most central RIS hypothesises that required to be tested to assert the relevancy of RIS for developing countries. The hypotheses were concerned with the importance of systemicness (H1); the link between TNC’s and the local innovation system, especially the local firms (H2); the importance of export and derived interaction with non-domestic users (H3): the centrality of universities (H4), and finally the role of state intervention and policy (H5). The hypotheses were tested on two regional innovation systems in Mexico: Jalisco and Baja California. The testing was based on original data and the development of a novel method for comparing systematically technological capabilities at firm and regional level. Earlier research on (regional) innovation systems in developing countries has emphasized the importance of adapting the conceptual and methodological apparatus to the specificities of the developing countries; special attention has been ascribed to the fact that developing countries need to rely on capital, technology and knowledge originating outside the boundaries of their home region/countries and thus the need to understand RIS in developing countries as open systems. The analysis shows that general claims in the RIS literature like systemicness (H1), centrality of universities (H4)xxxv and state intervention/policy (H5) all find support in the econometrical and qualitative analysis. In addition, the interaction between foreign subsidiaries and locally-owned firms is not significant to explain advanced technological capabilities in the two regions studied (H2) and the absence of localized users (H3) did not seem to hamper the innovative performance of the firms located in the two regional innovation systems. Hence, the overall findings of the paper suggests that the conceptual and related policy challenges associated with conceptualizing and constructing regional innovation systems in developing countries are smaller that assumed in the literature. Yet, our research needs to be complemented with other systematic econometrical empirical studies. Data are also

30

likely to reflect spatial-temporal specificities that needs to addressed; this call for larger comparative studies across time, industries and regions/countries. Finally, RIS research concerning developing countries – as it is also the case in our paper - needs to pay more attention to indigenous innovations originating outside formal knowledge creating industrial setting such as firms and universities. To our knowledge there has not been RIS studies concerned with, for example, innovations occurring in the informal economy or the rural communities. These innovations might not become institutionalized standard innovations but might hold important roles for reduction of poverty among the poor. Yet, we know nothing about the nature of these innovations, how they disseminate and which type of innovation systems that can support them. In this sense, innovation systems literature needs to start addressing the question of the direction of change, that is, what is being innovated and for whom instead of using innovation and standard economic performance measures (i.e. growth, competitiveness), to gain a stronger relevancy for coping with problems related to development. References Amin, A. (2004). "Regions unbound: towards a new politics and place." Geografiska

Annaler B 86(1): 31-42.

Archibugi, D. and B. Lundvall (2001). The Globalizing Learning Economy, Oxford University Press.

Archibugi, D. and C. Pietrobelli (2003). "The globalization of technology and its implications for developing countries - Windows of opportunity or further burden?" Technological Forecasting and Social Change 70(9): 861-883.

Ariffin, N. and P. Figueiredo (2003). "Internationalization of Innovative Capabilities:

counter-evidence from the electronics industry in Malaysia and Brazil." Oxford Development Studies 32(4): 559-583.

Arora, A. and A. Gambardella (2004). The globalization of the software industry: perspective and opportunities for developed and developing countries. NBER Working Paper Series.

Asheim B, Coenen L, Vang J, 2007, "Face-to-face, buzz, and knowledge bases: sociospatial implications for learning, innovation, and innovation policy" Environment and Planning C: Government and Policy 25(5) 655 – 670

Asheim, B. and A. Isaksen (2002). "Regional innovation system: the integration of local 'sticky' and global 'ubiquitous' knowledge." Journal of Technology Transfer 27: 77-86.

Asheim, B. and L. Coenen (2005). "Knowledge bases and regional innovation systems: Comparing Nordic clusters." Research Policy 34(8): 1173.

31

Asheim, B. and M. Gertler (2004). The geography of innovation: regional innovation systems. The Oxford Handbook of Innovation. J. Fagerberg, D. Mowery and R. Nelson. Oxford, OUP: 291-317.

Asheim, B., A. Isaksen, C. Nauwelaers and F. Toedtling (2003). Regional innovation policy for small-medium enterprises. Cheltenham, UK, MA: Edward Elgar.

Asheim, B., L. Coenen, J. Moodysson and J. Vang (2007). "Constructing knowledge-based regional advantage: implications for regional innovation policy." International Journal of Entrepreneurship and Innovation Management 7(2): 140-155.

Athreye, S. S. (2005). "The Indian software industry and its evolving service capability." Industrial and Corporate Change 14(3): 393-418.

Bagnasco, A. (1988)La Costruzione Sociale Del Mercato, Il Mulino, Bologna,

Baranson, J. and R. Roark (1985). "Trends in North-South Transfer of High Technology." International Technology Transfer: Concepts, Measures and Comparisons: 24-42.

Beccatini, G. (1990) The Marshallian industrial district as a socio-economic notion in F. Pyke, G. Becattini and W. Sengenberger (eds.) Industrial Districts and Inter-firm cooperation in Italy, Geneva, International Institute for Labor Statistics, pp. 37-51.

Bell, M. and K. Pavitt (1993). "Technological accumulation and industrial growth: contrasts between developed and developing countries." Industrial and Corporate Change 2(2): 157-210.

Bell, M. and K. Pavitt (1995). "The Development of Technological Capabilities." Trade, Technology and International Competitiveness: 69-101.

Birkinshaw, J. and N. Hood (1998). "Multinational Subsidiary Evolution: Capability and Charter Change in Foreign-Owned Subsidiary Companies." The Academy of Management Review 23(4): 773-795.

Brusco,S. (1982) The Emilian model: productive decentralisation and social integration, Cambridge Journal of Economics 1982,6, 167-184

Buitelaar, R., R. P. Pérez, R. Urrutia-Alvarez and U. N. E. D. O. P. Productivity and Management (2000). Costa Rica: Sistema Nacional de Innovación, Naciones Unidas, Comisión Económica para América Latina y el Caribe, División de Desarrollo Productivo y Empresarial.

Cantwell, J. and L. Piscitello (2002). "The location of technological activities of MNCs in European regions: the role of spillovers and local competencies." Journal of International Management 8(1): 69-96.

32

Carlsson, B. (2006). "Internationalization of innovation systems: A survey of the literature star, open." Research Policy 35(1): 56-67.

Carlsson, B., S. Jacobsson, M. Holmen and A. Rickne (2002). "Innovation systems: analytical and methodological issues." Research Policy 31: 233-245.

Casalet, M. (2000). "The Institutional Matrix and Its Main Functional Activities Supporting Innovation" in Developing Innovation Systems. Mexico in a Global Context. Londres y Nueva York: Continuum." Continuum, London, New York.

Castellacci, F. (2006). The interactions between national systems and sectoral patterns of innovation. TIK Working Papers. www.tik.uio.no

Caves, R. E. (1980). "Industrial organization, corporate strategy and structure." Journal of Economic Literature 58: 64-92.

Chaminade, C. and C. Edquist (2006). From theory to practice. The use of the systems of innovation approach in innovation policy. Innovation, Learning and Institutions. J. Hage and de Meeus, Oxford University Press.

Chaminade, C. and J. Vang (2006a). Globalisation of Knowledge Production and Regional Innovation Policy: Supporting Specialized Hubs in Developing Countries. CIRCLE Electronic Working Paper Series, CIRCLE: 38.

Chaminade, C. and J. Vang (2006b). Innovation Policy for Small and Medium Size SMEs in Asia: an Innovation Systems Perspective. Handbook of Research on Asian Business. H. W.-C. Yeung, Edward Elgar.

Chaminade, C. and J. Vang (Forthcoming). "Upgrading in Asian clusters: Rethinking the importance of interactive-learning." Science, Technology and Society.

Christensen, J. L. (1992). The role of finance in national systems of innovation. National systems of innovation and interactive learning. B.-A. Lundvall London, Pinter: 146-168.

Coenen, L. (2006). Faraway, so Close! The Changing Geographies of Regional Innovation. CIRCLE, Lund University. PhD.

Cohen, S. S. and G. Fields (1998). Social Capital and Capital Gains, Or Virtual Bowling in Silicon Valley, Berkeley Roundtable on the International Economy, University of California, Berkeley.

Cohen, W. and D. Levinthal (1990). "Absorptive capacity: a new perspective on learning and innovation." Administrative Science Quarterly 35: 128-152.

Cohen, W. M., R. R. Nelson and J. P. Walsh (2002). "Links and Impacts: The Influence of Public Research on Industrial R & D." Management Science 48(1): 1-23.

33

Cooke, P., M. Gomez-Uranga and G. Etxebarria (1997). "Regional systems of innovation: institutional and organizational dimensions." Research Policy 26: 475-491.

Dahlman, C. J. and P. J. Brimble (1990). Technology Strategy and Policy for Industrial Competitiveness: A Case Study in Thailand, World Bank Industry and Energy Dept., PRE.

Dalum, B., B. Johnson and B. Lundvall (1992). "Public Policy in the Learning Society." National Systems of Innovation. London: Pinter: 296-317.

Desai, A. (1985), “Indigenous and Foreign Determinants of Technological Change in Indian Industry”, Economic and Political Weekly,45: 2081-2094.

Dosi, G., K. Pavitt and L. Soete (1990). The economics of technical change and international trade, Harvester Wheatsheaf New York.

Dossani, R. and M. Kenney (2006). "Reflections upon" Sizing the Emerging Global Labor Market"." The Academy of Management Perspectives (formerly The Academy of Management Executive)(AMP) 20(4): 35-41.

Dunning, J. H. (1994). Reevaluating the Benefits of Foreign Direct Investment, University of Reading, Department of Economics.

Dutrénit, G. (2005). "Acumulación de capacidades tecnológicas en la industrica maquiladora." Comercio Exterior 55(7): 574-86.

Edquist, C., Ed. (1997). Systems of innovation: Technologies, institutions and organizations. London, Pinter.

Ernst, D. and L. Kim (2002). "global production networks, knwoledge diffusion and local capability formation." Research Policy 31: 1417-1429.

Etzkowitz, H. and L. Leydesdorff (2000). "The dynamics of innovation: from National Systems and “Mode 2” to a Triple Helix of university-industry-government relations." Research Policy 29(2): 109-123.

Evangelista, R., S. Iammarino, V. Mastrostefano and A. Silvani (2002). "Looking for Regional Systems of Innovation: Evidence from the Italian Innovation Survey." Regional Studies 36(2): 173-186.

Fagerberg, J. (2004). Innovation: a guide to the literature. The Oxford Handbook of Innovation. J. FAGERBERG, D. MOWERY and R. NELSON. Oxford, OUP: 1-29.

Freeman, C. (1987). Technology policy and economic performance: lessons from Japan. London, Pinter.

34

Galli, R. and M. Teubal (1997). Paradigmatic shifts in national innovation systems. systems of innovation - Technologies, Institutions and Organization. C. Edquist. London, Pinter.

Gertler, M. S. (2004). Manufacturing Culture: The Institutional Geography of Industrial Practice, Oxford University Press.

Giuliani, E. (2007). "The selective nature of knowledge networks in clusters: evidence from the wine industry." Journal of Economic Geography 7(2): 139.

Giuliani, E. and M. Bell (2005). "When micro shapes the meso: Learning networks in a Chilean wine cluster." Research Policy 34(1): 47-68.

Giuliani, E., R. Rabelotti and M. P. Van Dijk (2005). Clusters facing competition: the importance of external linkages. Aldershot and Burlington, Ashgate Publishing.

Gregersen, B. (1992), “The public sector as a pacer in national systems of innovation”, in B.A. Lundvall ed., National Systems of Innovation, Pinter, London: 129-145.

Grosman, G. and E. Helpman (1991). Innovation and Growth in the Global Economy,

Cambridge: MIT press.

Gunasekara, C. (2006) Reframing the role of universities in the development of regional innovation systems, Journal of technology transfer 31 (1) pp. 101-111.

Hayek, F. A. (1945). "The Use of Knowledge in Society." The American Economic Review 35(4): 519-530.

Hobday, M. (1995). Innovation in East Asia: The Challenge to Japan, Edward Elgar.

Howells, J. (1999). Regional systems of innovation. Innovation Policy in a Global Economy. D. Archibugi, J. Howells and J. Mitchie. Cambridge, Cambridge University Press: 67–93.

Humphrey, J. and H. Schmitz (2002). "How does insertion in global value chains affect upgrading in industrial clusters?" Regional Studies 36(9): 1017-1027.

Iammarino, S. (2005). "An evolutionary integrated view of regional innovation systems: concepts, measures and historical perspectives." European Planning Studies 13(4): 497–520.

Isaksen, A. (2003). Lock-in of Regional Clusters: The Case of Offshore Engineering in the Oslo region. Cooperation, Networks, and Institutions in Regional Innovation Systems. A. Isaksen. Cheltenham, Edward Elgar: 247-273.

Jaffe, A. B., M. Trajtenberg and R. Henderson (1993). "Geographic Localization of Knowledge Spillovers as Evidenced by Patent Citations." The Quarterly Journal of Economics 108(3): 577-598.

35

Jeppesen, L. B. and L. Frederiksen (2006). "Why Do Users Contribute to Firm-Hosted User Communities? The Case of Computer-Controlled Music Instruments." Organization Science 17(1): 45-63.

Jeppesen, L.B. and Molin, M.J., (2003) Consumers as Co-developers: Learning and Innovation Outside the Firm. Technology Analysis & Strategic Management15 (3) 363-84.

Kaplinsky, R. (2005). Globalization, Poverty and Inequality: Between a Rock and a Hard Place, Polity Press.

Kim, S. R. and N. Von Tunzelmann (1998). Aligning Internal and External Networks: Taiwan's Specialization in IT, University of Sussex, SPRU: Science and Technology Policy Research.

Kline, S. and N. Rosenberg (1986). An overview of innovation. The positive sum strategy. L. a. ROSENBERG. Washington D:C, National Academy of Sciences: 289.

Lall, S. (1992), “Technological Capabilities and Industrialization”, World Development, 20 (2): 165-186.

Loebis, L. and H. Schmitz (2005). Java furniture makers: winners or losers from globalisation? Development and practice, IDS, Sussex: 10.

Long, J.S. (1997), Regression Models for Categorical and Limited Dependent Variables, SAGE Publications, Thousand Oaks, California.

Luhmann, N. (1995). Social systems, Stanford University Press Stanford, California.

Lundvall, B. A. and S. Borrás (1999). "The Globalising Learning Economy: Implications for Innovation Policy." Office for Official Publications of the European Communities, Luxembourg.

Lundvall, B.-A. (1988). Innovation as an interactive process: from user-producer interaction to the national system of innovation. Technical change and economic theory. G. e. a. DOSI. London, Pinter.

Lundvall, B.-A. and S. Borrás (2004). Science, Technology and Innovation Policy. The Oxford Hanbook of Innovation. J. Fagerberg, D. Mowery and R. Nelson. Oxford, Oxford University Press.

Lundvall, B.-A., 2002. The University in the Learning Economy, DRUID Working Papers, No. 6, ISBN: 87-7873-122-4.

Lundvall, B.-A., Ed. (1992). National systems of innovation. Towards a theory of innovation and interactive learning. Londres, Pinter.

36

Lundvall, B.-A., P. Intarakumnerd and J. Vang (2006). Asian Innovation Systems in Transition, Edward Elgar.

Luthria, M. and I. Nabi (2002). Building Competitive Firms: Incentives and Capabilities, World Bank Publications.

Mansfield, E. and J.Y. Lee (1996), “The modern university: contributor to industrial innovation and recipient of industrial R&D support”, Research Policy, 25 (7): 1047-1058.

Martin, R. and P. Sunley (2003). "Deconstructing clusters: chaotic concept or policy panacea?" Journal of Economic Geography(3): 5-35.

Mathews, J (2000) Tiger Technology. The Creation of the Semiconductor Industry in East Asia (with Dong-Sung Cho). Cambridge: Cambridge University Press

Mathews, J (2002) Dragon Multinational: A New Model for Global Growth, New York: Oxford University Press

Mowery, D. (1995). The practice of technology policy. Handbook of the economics of innovation and technological change. P. Stoneman. Oxford, Blackwell: 513-557.

Mowery, D. and B. N. Sampat (2004). Universities in national innovation systems. The Oxford Handbook of Innovation. J. FAGERBERG, D. MOWERY and R. NELSON. Oxford, OUP: 209-239.

Narula, R. (2001). Multinational Firms, Regional Integration and Globalising Markets: Implications for Developing Countries. Trade and Regional Integration in teh Development Agenda. R. Devlin and A. Estevadeordal. Washington DC, Brookings Institution.

Nelson, R. and S. Winter (1982). An evolutionay theory of economic change. Cambridge-MA, Harvard University Press.

Nelson, R., Ed. (1993). National innovation systems. A comparative analysis. Nueva York, Oxford Univ. Press.

O’donnell, S. and T. Blumentritt (1999). "The contribution of foreign subsidiaries to host country national competitiveness." Journal of International Management 5(3): 187-206.

O’Sullivan, M. (2005), “Finance and innovation”, in J. Fagerberg et al., The Oxford Handbook of Innovation, Oxford University Press, Oxford: 240-265.

Padilla-Perez, R. (2006). Foreign direct investment and regional technological capabilities: the case of the electronics industry in Mexico. SPRU: Science Policy Research Unit, Sussex University. PhD.

37

Pietrobelli, C. and R. Rabellotti (2004). Upgrading in clusters and value chains in LatinAmerica: the role of policies. Sustainable department Best Practices Series. New York, Inter-American Development Bank: 97.

Pietrobelli, C. and R. Rabellotti (2006). Upgrading to Compete: Global Value Chains, Clusters, and SMEs in Latin America, Inter-American Development Bank; David Rockefeller Center for Latin American Studies, Harvard University.

Piore, M. and C. Sabel (1984). The second industrial divide. New York, Basic Books.

Porter, M. E. (1998). "Clusters and the new economics of competition." Harvard Business Review 76(6): 77-90.

Rodriguez-Pose, A. and M. Storper (2006). "Better rules or stronger communities? On the social foundations of institutional change and its economic effects." Economic Geography 82(1): 1–25.

Romijn, H. (1999). Acquisition of Technological Capability in Small Firms in Developing Countries, St. Martin's Press New York.

Sardana, D. and V. V. Krishna (2006). "Government, University and Industry Relations: The Case of Biotechnology in the Delhi Region." Science, Technology and Society 11(2): 351.

Saxenian, A. L. (2006). The New Argonauts: Regional Advantage in a Global Economy, Harvard University Press.

Saxenian, A.-L. (1994). Regional advantage: culture and ocmpetition in Silicon Valley and Route 128. Cambridge, Harvard University Press.

Schmitz, H., 2006. Regional systems and global chains. Paper presented at the Fifth International Conference on Industrial Clustering and Regional Development. oec.pku.edu.cn/icrd/

Scott, A. J. and G. Garofoli (2007). Development on the Ground: Clusters, networks and regions in emerging economies, Oxford Univ Press.

Soderberg, A. M. and N. Holden (2002). "Rethinking Cross Cultural Management in a Globalizing Business World." International Journal of Cross Cultural Management 2(1): 103.

Staber, U. (1996). "Accounting for variations in the performance of industrial districts: the case of Baden-Wuttemberg." International Journal of Urban and Regional Research 20: 299-316.

Storper, M., Venables. A. J., 2004. Buzz: face- to-face contact and the urban Economy. Journal of Economic Geography 4, 351–70.

38

Trigilia, C.: 1990, “Work and Politics in the Third Italy’s Industrial Districts”, in F. Pyke, G. Becattini and W. Sengenberger (eds.), Industrial Districts and Interfirm Co-operation in Italy (Geneva: International Institute for Labour Studies), pp. 160–184.

Turpin, T. and C. Martinez-Fernandez (2003). "Riding the waves of policy." Science, Technology and Society 8(2): 215–234.

Vang, J. and B. Asheim (2006). "Regions, Absorptive Capacity and Strategic Coupling with High-Tech TNCs: Lessons from India and China." Science Technology & Society 11(1): 39.

Vang, J. and C. Chaminade (2006). Polices for Building Bangalore´s IT Future: An Innovation Systems Perspective, CIRCLE.

Vang, J. and M. Overby (2006). Transnational communities, TNCs and Development: The case for the Indian IT services industry. Asian Innovation Systems in Transition. B.-A. Lundvall, I. Patarapong and J. Vang, Edward Elgar.

Vang, J., C. Chaminade and L. Coenen (Forthcoming). Learning from the Bangalore Experience: The Role of Universities in an Emerging Regional Innovation System. New Asian Dynamics of Science, Technology and Innovation (ST&I). Series in "Technology, Globalization and Development”. A. D’Costa and G. Parayil, Palgrave Macmillan.

Viotti, E. B. (2002). "National Learning Systems-A new approach on technological change in late industrializing economies and evidences from the cases of Brazil and South Korea." Technological Forecasting and Social Change 69(7): 653-680.

Von Hippel, E. (1988). Sources of innovation. Oxford, Oxford University Press.

Von Tunzelmann, N. and Q. Wang (2001). "Resources and capabilities in a context of radical change." Conference in Honour of Edith Penrose, INSEAD, Fontainebleau.

Von Tunzelmann, G.N. (2006), Regional capabilities and industrial regeneration, mimeo, SPRU, University of Sussex.

Yeung, H. W.-C. (2006). Situating Regional Development in the competitive dynamics of

global production networks: an east asian perspective. Working Paper Series: 47.

Young, S., N. Hood and E. Peters (1994). "Multinational Enterprises and Regional Economic Development." Regional Studies 28(7): 657-77.

Zanfei, A. (2005). "Globalization at bay? Multinational growth and technology spillover." Critical perspectives on International Business 1(1): 5-17.

39

Appendix A.1 TECHNOLOGY CAPABILITIES INDICES TC PROCESS Ordinal variable, maximum value = 3, minimum value = 1 TC proc = 1 if firm i had not modified or adapted machinery and equipment, or

had only carried out minor adaptations to the local conditions; did not operate under advanced management techniques; and had not been certified by internal standards

TC proc = 2 only if firm i fulfilled ALL the following: operated under advanced management techniques (at least 3 out of 5 techniques listed in the questionnaire); had been certified by internal standards; was characterized by flexible production schemes; and had modified machinery and equipment to increase efficiency

TC proc = 3 only if firm i fulfilled ALL the requirements in the above level (TCproc = 2) and additionally had developed new equipment and software

TC PRODUCT Ordinal variable, maximum value = 3, minimum value = 1 TC prod = 1 if firm i received product specification from the parent company or

clients and had not carried out production adaptation and modification

TC prod = 2 if firm i had a product design department (design for manufacturability) and had frequently modified and improved its products

TC prod = 3 if firm i had carried out R&D activities and had developed new or significantly improved products

TC OVERALL = TCproc + TCprod

Ordinal variable, maximum value = 6, minimum value = 2

i The large regional inequalities in countries like India are noteworthy. While certain Indian regions display high growth rates others maintain a level of economic development comparable to selected regions in Sub-Sahara Africa. ii For example, South Korea has employed a state-centered model relying on a flexible ‘penduling’ between import-substitution industrialisation (ISI) and export-oriented strategies for industrialization and even becoming industry-leaders in selected fields. Singapore, China and India have relied on FDI for their development, upgrading and innovation strategies; yet approached their home markets and applied strategies for constructing indigenous capabilities in a variety of different ways. Mexico has chosen to focus on exploiting their physical proximity to the US, and so forth. iii Carlsson et al. (2002, p. 243) define a system as “a set of interrelated components” (i.e. made up of components, relationships and attributes). iv Firms, of course, not only exploit knowledge but also develop new knowledge. v RIS has not yet theorized the conflicts and mutuality between informal institutions (i.e. norms and values) and formal rules (i.e. society), see Rodríguez-Pose and Storper 2006. RIS has also neglected – in line with innovation systems research in general – to engage in dialog with more recent research on institutions and cultures suggesting that culture cannot be reduced to norms and values but should be seen as the dynamic result of ongoing negotiations (Søderberg and Holden 2002). vi However, the empirical support for this thesis on proximity and interactive learning is contested. While Jaffe et al. (1993) for example, find support for knowledge spillovers within a certain regional innovation system, other more recent studies emphasize the unequal nature of localized learning in clusters (Giuliani , 2007) and the importance of absorptive capacity. vii To a very minor extent the indigenous firms ability to tap into overseas knowledge hubs (Mathews 2000, 2002, Vang and Overby 2006, Saxenian 2006).

40

viii The notion used is also at odds with Hayek’s (1945) notion of spontaneous self-organizing systems (i.e. catallaxy). ix For a more detailed discussion on the interaction between local and national levels of policy making see Vang and Chaminade (2006).. x The role of the state supporting innovation is highly contested in developing countries. As some research shows, the state might even aggravate the systemic problems, through for example, the development of an inadequate institutional framework (or the absence of it), adverse selection mechanisms or even competing with the private actors to access scarce resources. This suggests that states – regional or national – and policy should not always be considered constitutive elements in creation of RIS in developing countries. For example, several empirical studies of Bangalore have suggested that there has not been a need for state intervention (apart from education policy) in at least the early phases of the development of the RIS (Athreye, 2005; Arora and Gambardella, 2004). xi See, for instance, Caves (1980); Grossman and Helpman (1991); Dunning (1994); xii See Padilla-Perez (2006) for further details on this methodology. xiii It is important to acknowledge that, this regional matrix was developed to study an FDI-led, technology-intensive industry in a developing country. The taxonomy was created on the basis of the existing literature, as recognized below, and our own fieldwork. xiv Competences are understood as inputs to produce goods and services, and capabilities involve contemporaneous learning and the accumulation of new knowledge, and the integration of behavioral, social and economic factors. See von Tunzelmann and Wang (2003). xv Several authors have studied technological capabilities at firm level, using different classifications: production, investment, innovation, operation, acquisition, linkage, etc. In general, these classifications aim at decomposing the constituent elements of technological activity within the firm. See, for instance, Desai (1985), Baranson and Roark (1985), Dahlman and Brimble (1990), Lall (1992), Bell and Pavitt (1995), Kim and von Tunzelmann (1998), Romijn (1999) and Viotti (2002). The classification here aims on the one hand to simplify the analysis, and on the other to distinguish between competences and capabilities. xvi The classification of technological capabilities into three levels was used by Lall (1992), Bell and Pavitt (1995) and Ariffin and Figuereido (2003). xvii See Padilla-Perez (2005) for further information on the electronics industry. xviii There is a myriad of studies on technical change within the firm. Some of the references to choose the factors potentially associated with firm-level technological capabilities are: Nelson & Winter (1982); Dosi, Pavitt & Soete (1990); Freeman & Soete (1997) and Romijn (1999). xix See Dalum et al. (1992); Gregersen (1992); Mowery (1995); Freeman and Soete (1997); Dutrénit (2005), and Lundvall and Borrás (2004). xx For more information, see Buitelaar, Padilla-Pérez and Urrutia-Alvarez (2000); and Casalet (2000). xxi See Padilla-Perez (2006) for further information on how the population was identified and the sample constructed. xxii Following Ernst & Kim (2002), four types of firms can be identified in the electronics industry, each with different technological characteristics: original equipment manufacturers, contract manufacturers, suppliers and design houses. The second criterion was intended to give a representative sample of foreign-owned firms and locally-owned firms and, within the former, to cover firms from different nationalities. xxiii See Long (1997) for more information on ordinal variables. xxiv CANIETI Occidente is a regional office of the national industrial association for the electronics industry. CADELEC (Cadena Productiva de la Electrónica) is a private organization that provides administrative, technological and logistic support to firms active in the electronics industry and their suppliers. xxv Along the same lines, Birkinshaw and Hood (1998) argue that decisions made by subsidiary managers regarding the activities undertaken by the subsidiary are crucial to explaining subsidiary evolution. xxvi The regressions for foreign subsidiaries have 53 observations, and 27 for locally-owned firms. xxvii The correlation among independent variables is higher than in the whole sample. To prevent multicolinearity, this final specification does not include highly correlated variables. xxviii Only the results that are relevant for the analysis are reported. xxix Indirect goods are not directly incorporated in the final good, for example: packing and wrapping products, furniture, consumable goods, labels, bags, foam, fabrics, gloves, cleaning products and paper board.

41

xxx Size was significant and positive for overall and process technological capabilities. xxxi Other factors like size, human capital (unqualified/qualified and direct/indirect) and growth were also significantly associated with technological capabilities. The detailed results can be found in Padilla Perez (2006). xxxii It is important to mention that although the electronics industry is larger in Baja California, this state is much less densely populated than Jalisco. xxxiii Available on line http://www.bajacalifornia.gob.mx/sedeco/web_2005/estadisticas_bc.htm xxxiv Research and Advanced Studies Centre. xxxv It should be noted that what finds support is the conceptualization of developmental university; not the tested hypothesis.

CIRCLE ELECTRONIC WORKING PAPERS SERIES (EWP) CIRCLE (Centre for Innovation, Research and Competence in the Learning Economy) is a multidisciplinary research centre set off by several faculties at Lund University and Blekinge Institute of Technology. CIRCLE has a mandate to conduct multidisciplinary research and education on the following issues: Long-term perspectives on innovation, structural change and economic growth, Entrepreneurship and venture capital formation with a special focus on new ventures, The dynamics of R&D systems and technological systems, including their impact on entrepreneurship and growth, Regional innovation systems in different national and international contexts, International comparative analyses of national innovation systems and Policy design based on policy learning, The CIRCLE Electronic Working Paper Series are intended to be an instrument for early dissemination of the research undertaken by CIRCLE researchers, associates and visiting scholars and stimulate discussion and critical comment. The working papers present research results that in whole or in part are suitable for submission to a refereed journal or to the editor of a book or have already been submitted and/or accepted for publication. CIRCLE EWPs are available on-line at: http://www.circle.lu.se/publications Available papers: 2008 WP 2008/01 R&D and financial systems: the determinants of R&D expenditures in the Swedish pharmaceutical industry Malmberg, Claes WP 2008/02 The Development of a New Swedish Innovation Policy. A Historical Institutional Approach Persson, Bo WP 2008/03 The Effects of R&D on Regional Invention and Innovation Olof Ejermo and Urban Gråsjö WP 2008/04 Clusters in Time and Space: Understanding the Growth and Transformation of Life Science in Scania Moodysson, Jerker; Nilsson, Magnus; Svensson Henning, Martin WP 2008/05 Building absorptive capacity in less developed countries The case of Tanzania Szogs, Astrid; Chaminade, Cristina and Azatyan, Ruzana WP 2008/06 Design of Innovation Policy through Diagnostic Analysis: Identification of Systemic Problems (or Failures) Edquist, Charles

WP 2008/07 The Swedish Paradox arises in Fast-Growing Sectors Ejermo, Olof; Kander, Astrid and Svensson Henning, Martin

WP 2008/08 Policy Reforms, New University-Industry Links and Implications for Regional Development in Japan Kitagawa, Fumi

WP 2008/09 The Challenges of Globalisation: Strategic Choices for Innovation Policy Borrás, Susana; Chaminade, Cristina and Edquist, Charles

WP 2008/10 Comparing national systems of innovation in Asia and Europe: theory and comparative framework Edquist, Charles and Hommen, Leif

WP 2008/11 Putting Constructed Regional Advantage into Swedish Practice? The case of the VINNVÄXT initiative 'Food Innovation at Interfaces' Coenen, Lars; Moodysson, Jerker

WP 2008/12 Energy transitions in Europe: 1600-2000 Kander, Astrid; Malanima, Paolo and Warde, Paul WP 2008/13 RIS and Developing Countries: Linking firm technological capabilities to regional systems of innovation Padilla, Ramon; Vang, Jan and Chaminade, Cristina 2007 WP 2007/01 Path-following or Leapfrogging in Catching-up: the Case of Chinese Telecommunication Equipment Industry Liu, Xielin WP 2007/02 The effects of institutional change on innovation and productivity growth in the Swedish pharmaceutical industry Malmberg, Claes WP 2007/03 Global-local linkages, Spillovers and Cultural Clusters: Theoretical and Empirical insights from an exploratory study of Toronto’s Film Cluster Vang, Jan; Chaminade, Cristina

WP 2007/04 Learning from the Bangalore Experience: The Role of Universities in an Emerging Regional Innovation System Vang, Jan; Chaminade, Cristina.; Coenen, Lars. WP 2007/05 Industrial dynamics and innovative pressure on energy -Sweden with European and Global outlooks Schön, Lennart; Kander, Astrid. WP 2007/06 In defence of electricity as a general purpose technology Kander, Astrid; Enflo, Kerstin; Schön, Lennart WP 2007/07 Swedish business research productivity – improvements against international trends Ejermo, Olof; Kander, Astrid WP 2007/08 Regional innovation measured by patent data – does quality matter? Ejermo, Olof WP 2007/09 Innovation System Policies in Less Successful Developing countries: The case of Thailand Intarakumnerd, Patarapong; Chaminade, Cristina 2006 WP 2006/01 The Swedish Paradox Ejermo, Olof; Kander, Astrid WP 2006/02 Building RIS in Developing Countries: Policy Lessons from Bangalore, India Vang, Jan; Chaminade, Cristina WP 2006/03 Innovation Policy for Asian SMEs: Exploring cluster differences Chaminade, Cristina; Vang, Jan. WP 2006/04 Rationales for public intervention from a system of innovation approach: the case of VINNOVA. Chaminade, Cristina; Edquist, Charles WP 2006/05 Technology and Trade: an analysis of technology specialization and export flows Andersson, Martin; Ejermo, Olof

WP 2006/06 A Knowledge-based Categorization of Research-based Spin-off Creation Gabrielsson, Jonas; Landström, Hans; Brunsnes, E. Thomas WP2006/07 Board control and corporate innovation: an empirical study of small technology-based firms Gabrielsson, Jonas; Politis, Diamanto WP2006/08 On and Off the Beaten Path: Transferring Knowledge through Formal and Informal Networks Rick Aalbers; Otto Koppius; Wilfred Dolfsma WP2006/09 Trends in R&D, innovation and productivity in Sweden 1985-2002 Ejermo, Olof; Kander, Astrid WP2006/10 Development Blocks and the Second Industrial Revolution, Sweden 1900-1974 Enflo, Kerstin; Kander, Astrid; Schön, Lennart WP 2006/11 The uneven and selective nature of cluster knowledge networks: evidence from the wine industry Giuliani, Elisa WP 2006/12 Informal investors and value added: The contribution of investors’ experientially acquired resources in the entrepreneurial process Politis, Diamanto; Gabrielsson, Jonas WP 2006/13 Informal investors and value added: What do we know and where do we go? Politis, Diamanto; Gabrielsson, Jonas WP 2006/14 Inventive and innovative activity over time and geographical space: the case of Sweden Ejermo, Olof 2005 WP 2005/1 Constructing Regional Advantage at the Northern Edge Coenen, Lars; Asheim, Bjørn WP 2005/02 From Theory to Practice: The Use of the Systems of Innovation Approach for Innovation Policy Chaminade, Cristina; Edquist, Charles WP 2005/03 The Role of Regional Innovation Systems in a Globalising Economy: Comparing Knowledge Bases and Institutional Frameworks in Nordic Clusters

Asheim, Bjørn; Coenen, Lars WP 2005/04 How does Accessibility to Knowledge Sources Affect the Innovativeness of Corporations? Evidence from Sweden Andersson, Martin; Ejermo, Olof WP 2005/05 Contextualizing Regional Innovation Systems in a Globalizing Learning Economy: On Knowledge Bases and Institutional Frameworks Asheim, Bjørn; Coenen, Lars WP 2005/06 Innovation Policies for Asian SMEs: An Innovation Systems Perspective Chaminade, Cristina; Vang, Jan WP 2005/07 Re-norming the Science-Society Relation Jacob, Merle WP 2005/08 Corporate innovation and competitive environment Huse, Morten; Neubaum, Donald O.; Gabrielsson, Jonas WP 2005/09 Knowledge and accountability: Outside directors' contribution in the corporate value chain Huse, Morten, Gabrielsson, Jonas; Minichilli, Alessandro WP 2005/10 Rethinking the Spatial Organization of Creative Industries Vang, Jan WP 2005/11 Interregional Inventor Networks as Studied by Patent Co-inventorships Ejermo, Olof; Karlsson, Charlie WP 2005/12 Knowledge Bases and Spatial Patterns of Collaboration: Comparing the Pharma and Agro-Food Bioregions Scania and Saskatoon Coenen, Lars; Moodysson, Jerker; Ryan, Camille; Asheim, Bjørn; Phillips, Peter WP 2005/13 Regional Innovation System Policy: a Knowledge-based Approach Asheim, Bjørn; Coenen, Lars; Moodysson, Jerker; Vang, Jan WP 2005/14 Face-to-Face, Buzz and Knowledge Bases: Socio-spatial implications for learning and innovation policy Asheim, Bjørn; Coenen, Lars, Vang, Jan WP 2005/15 The Creative Class and Regional Growth: Towards a Knowledge Based Approach Kalsø Hansen, Høgni; Vang, Jan; Bjørn T. Asheim

WP 2005/16 Emergence and Growth of Mjärdevi Science Park in Linköping, Sweden Hommen, Leif; Doloreux, David; Larsson, Emma WP 2005/17 Trademark Statistics as Innovation Indicators? – A Micro Study Malmberg, Claes